diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_1/README.md b/fed-learning/non-dp/tuned-llama-guard3-round_1/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e544b712adee647f829c2ca8c4571ab05915547a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_1/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Llama-Guard-3-8B
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_1/adapter_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_1/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..329293a4ef5d7c22bff7ad8a59dc4b7622ffc430
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_1/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a1ab765d5bfa0b7d3c35ff748fb8f1813a8150d13a764a19cb6acd2f61d77de
+size 810
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_1/adapter_model.safetensors b/fed-learning/non-dp/tuned-llama-guard3-round_1/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..5e4a226d789763aea74161975bc75ac63a48c14b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_1/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:54025514bb97586dd0dd8f84772d1f6ec6c48980d6b8c7f929355c9ef8eb5b99
+size 27297032
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_1/chat_template.jinja b/fed-learning/non-dp/tuned-llama-guard3-round_1/chat_template.jinja
new file mode 100644
index 0000000000000000000000000000000000000000..8e734bc48af8bcfeecc2d0d1736179ed81008826
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_1/chat_template.jinja
@@ -0,0 +1,30 @@
+{% if messages|length % 2 == 0 %}{% set role = 'Agent' %}{% else %}{% set role = 'User' %}{% endif %}{{ " <|begin_of_text|><|start_header_id|>user<|end_header_id|>
+
+Task: Check if there is unsafe content in '" + role + "' messages in conversations according our safety policy with the below categories.
+
+<BEGIN UNSAFE CONTENT CATEGORIES>
+S1: Violent Crimes.
+S2: Non-Violent Crimes.
+S3: Sex Crimes.
+S4: Child Exploitation.
+S5: Defamation.
+S6: Specialized Advice.
+S7: Privacy.
+S8: Intellectual Property.
+S9: Indiscriminate Weapons.
+S10: Hate.
+S11: Self-Harm.
+S12: Sexual Content.
+S13: Elections.
+S14: Code Interpreter Abuse.
+<END UNSAFE CONTENT CATEGORIES>
+
+<BEGIN CONVERSATION>
+
+" }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{% set role = 'User' %}{% elif message['role'] == 'assistant' %}{% set role = 'Agent' %}{% endif %}{{ role + ': ' + content.strip() + '
+
+' }}{% endfor %}{{ "<END CONVERSATION>
+
+Provide your safety assessment for ONLY THE LAST " + role + " message in the above conversation:
+ - First line must read 'safe' or 'unsafe'.
+ - If unsafe, a second line must include a comma-separated list of violated categories.<|eot_id|><|start_header_id|>assistant<|end_header_id|>" }}
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_1/special_tokens_map.json b/fed-learning/non-dp/tuned-llama-guard3-round_1/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..92d7fab6729d723f45689803b99006fec945cc47
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_1/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b1835caa5b4d70acaa210fa222b0036f1882f9525c4660fd4810fb3e1e40ff8
+size 325
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_1/tokenizer.json b/fed-learning/non-dp/tuned-llama-guard3-round_1/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af2d22fff23798bea98b1730ae7cdacaee0b087a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_1/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ade1dac458f86f9bea8bf35b713f14e1bbed24228429534038e9f7e54ea3e8b6
+size 17208712
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_1/tokenizer_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_1/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..32408241ddf390e43ec24f8cac8608a7c66f9d1f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_1/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc2a3cb7e7ae6678f28b4a1faee31a00e0f75c475a0fd6af06a100c7f0cde0e5
+size 50550
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_10/README.md b/fed-learning/non-dp/tuned-llama-guard3-round_10/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e544b712adee647f829c2ca8c4571ab05915547a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_10/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Llama-Guard-3-8B
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_10/adapter_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_10/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..329293a4ef5d7c22bff7ad8a59dc4b7622ffc430
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_10/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a1ab765d5bfa0b7d3c35ff748fb8f1813a8150d13a764a19cb6acd2f61d77de
+size 810
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_10/adapter_model.safetensors b/fed-learning/non-dp/tuned-llama-guard3-round_10/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..553bb9d9145f7e77455bb0b76f1f25e45bf228da
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_10/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f55528c840e8f0037a342825e95c41b3c7935402adf91f7372f8e7c9b1885dc5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_10/chat_template.jinja b/fed-learning/non-dp/tuned-llama-guard3-round_10/chat_template.jinja
new file mode 100644
index 0000000000000000000000000000000000000000..8e734bc48af8bcfeecc2d0d1736179ed81008826
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_10/chat_template.jinja
@@ -0,0 +1,30 @@
+{% if messages|length % 2 == 0 %}{% set role = 'Agent' %}{% else %}{% set role = 'User' %}{% endif %}{{ " <|begin_of_text|><|start_header_id|>user<|end_header_id|>
+
+Task: Check if there is unsafe content in '" + role + "' messages in conversations according our safety policy with the below categories.
+
+<BEGIN UNSAFE CONTENT CATEGORIES>
+S1: Violent Crimes.
+S2: Non-Violent Crimes.
+S3: Sex Crimes.
+S4: Child Exploitation.
+S5: Defamation.
+S6: Specialized Advice.
+S7: Privacy.
+S8: Intellectual Property.
+S9: Indiscriminate Weapons.
+S10: Hate.
+S11: Self-Harm.
+S12: Sexual Content.
+S13: Elections.
+S14: Code Interpreter Abuse.
+<END UNSAFE CONTENT CATEGORIES>
+
+<BEGIN CONVERSATION>
+
+" }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{% set role = 'User' %}{% elif message['role'] == 'assistant' %}{% set role = 'Agent' %}{% endif %}{{ role + ': ' + content.strip() + '
+
+' }}{% endfor %}{{ "<END CONVERSATION>
+
+Provide your safety assessment for ONLY THE LAST " + role + " message in the above conversation:
+ - First line must read 'safe' or 'unsafe'.
+ - If unsafe, a second line must include a comma-separated list of violated categories.<|eot_id|><|start_header_id|>assistant<|end_header_id|>" }}
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_10/special_tokens_map.json b/fed-learning/non-dp/tuned-llama-guard3-round_10/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..92d7fab6729d723f45689803b99006fec945cc47
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_10/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b1835caa5b4d70acaa210fa222b0036f1882f9525c4660fd4810fb3e1e40ff8
+size 325
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_10/tokenizer.json b/fed-learning/non-dp/tuned-llama-guard3-round_10/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af2d22fff23798bea98b1730ae7cdacaee0b087a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_10/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ade1dac458f86f9bea8bf35b713f14e1bbed24228429534038e9f7e54ea3e8b6
+size 17208712
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_10/tokenizer_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_10/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..32408241ddf390e43ec24f8cac8608a7c66f9d1f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_10/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc2a3cb7e7ae6678f28b4a1faee31a00e0f75c475a0fd6af06a100c7f0cde0e5
+size 50550
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_11/README.md b/fed-learning/non-dp/tuned-llama-guard3-round_11/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e544b712adee647f829c2ca8c4571ab05915547a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_11/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Llama-Guard-3-8B
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_11/adapter_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_11/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..329293a4ef5d7c22bff7ad8a59dc4b7622ffc430
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_11/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a1ab765d5bfa0b7d3c35ff748fb8f1813a8150d13a764a19cb6acd2f61d77de
+size 810
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_11/adapter_model.safetensors b/fed-learning/non-dp/tuned-llama-guard3-round_11/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..553bb9d9145f7e77455bb0b76f1f25e45bf228da
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_11/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f55528c840e8f0037a342825e95c41b3c7935402adf91f7372f8e7c9b1885dc5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_11/chat_template.jinja b/fed-learning/non-dp/tuned-llama-guard3-round_11/chat_template.jinja
new file mode 100644
index 0000000000000000000000000000000000000000..8e734bc48af8bcfeecc2d0d1736179ed81008826
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_11/chat_template.jinja
@@ -0,0 +1,30 @@
+{% if messages|length % 2 == 0 %}{% set role = 'Agent' %}{% else %}{% set role = 'User' %}{% endif %}{{ " <|begin_of_text|><|start_header_id|>user<|end_header_id|>
+
+Task: Check if there is unsafe content in '" + role + "' messages in conversations according our safety policy with the below categories.
+
+<BEGIN UNSAFE CONTENT CATEGORIES>
+S1: Violent Crimes.
+S2: Non-Violent Crimes.
+S3: Sex Crimes.
+S4: Child Exploitation.
+S5: Defamation.
+S6: Specialized Advice.
+S7: Privacy.
+S8: Intellectual Property.
+S9: Indiscriminate Weapons.
+S10: Hate.
+S11: Self-Harm.
+S12: Sexual Content.
+S13: Elections.
+S14: Code Interpreter Abuse.
+<END UNSAFE CONTENT CATEGORIES>
+
+<BEGIN CONVERSATION>
+
+" }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{% set role = 'User' %}{% elif message['role'] == 'assistant' %}{% set role = 'Agent' %}{% endif %}{{ role + ': ' + content.strip() + '
+
+' }}{% endfor %}{{ "<END CONVERSATION>
+
+Provide your safety assessment for ONLY THE LAST " + role + " message in the above conversation:
+ - First line must read 'safe' or 'unsafe'.
+ - If unsafe, a second line must include a comma-separated list of violated categories.<|eot_id|><|start_header_id|>assistant<|end_header_id|>" }}
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_11/special_tokens_map.json b/fed-learning/non-dp/tuned-llama-guard3-round_11/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..92d7fab6729d723f45689803b99006fec945cc47
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_11/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b1835caa5b4d70acaa210fa222b0036f1882f9525c4660fd4810fb3e1e40ff8
+size 325
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_11/tokenizer.json b/fed-learning/non-dp/tuned-llama-guard3-round_11/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af2d22fff23798bea98b1730ae7cdacaee0b087a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_11/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ade1dac458f86f9bea8bf35b713f14e1bbed24228429534038e9f7e54ea3e8b6
+size 17208712
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_11/tokenizer_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_11/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..32408241ddf390e43ec24f8cac8608a7c66f9d1f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_11/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc2a3cb7e7ae6678f28b4a1faee31a00e0f75c475a0fd6af06a100c7f0cde0e5
+size 50550
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_12/README.md b/fed-learning/non-dp/tuned-llama-guard3-round_12/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e544b712adee647f829c2ca8c4571ab05915547a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_12/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Llama-Guard-3-8B
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_12/adapter_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_12/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..329293a4ef5d7c22bff7ad8a59dc4b7622ffc430
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_12/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a1ab765d5bfa0b7d3c35ff748fb8f1813a8150d13a764a19cb6acd2f61d77de
+size 810
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_12/adapter_model.safetensors b/fed-learning/non-dp/tuned-llama-guard3-round_12/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..553bb9d9145f7e77455bb0b76f1f25e45bf228da
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_12/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f55528c840e8f0037a342825e95c41b3c7935402adf91f7372f8e7c9b1885dc5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_12/chat_template.jinja b/fed-learning/non-dp/tuned-llama-guard3-round_12/chat_template.jinja
new file mode 100644
index 0000000000000000000000000000000000000000..8e734bc48af8bcfeecc2d0d1736179ed81008826
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_12/chat_template.jinja
@@ -0,0 +1,30 @@
+{% if messages|length % 2 == 0 %}{% set role = 'Agent' %}{% else %}{% set role = 'User' %}{% endif %}{{ " <|begin_of_text|><|start_header_id|>user<|end_header_id|>
+
+Task: Check if there is unsafe content in '" + role + "' messages in conversations according our safety policy with the below categories.
+
+<BEGIN UNSAFE CONTENT CATEGORIES>
+S1: Violent Crimes.
+S2: Non-Violent Crimes.
+S3: Sex Crimes.
+S4: Child Exploitation.
+S5: Defamation.
+S6: Specialized Advice.
+S7: Privacy.
+S8: Intellectual Property.
+S9: Indiscriminate Weapons.
+S10: Hate.
+S11: Self-Harm.
+S12: Sexual Content.
+S13: Elections.
+S14: Code Interpreter Abuse.
+<END UNSAFE CONTENT CATEGORIES>
+
+<BEGIN CONVERSATION>
+
+" }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{% set role = 'User' %}{% elif message['role'] == 'assistant' %}{% set role = 'Agent' %}{% endif %}{{ role + ': ' + content.strip() + '
+
+' }}{% endfor %}{{ "<END CONVERSATION>
+
+Provide your safety assessment for ONLY THE LAST " + role + " message in the above conversation:
+ - First line must read 'safe' or 'unsafe'.
+ - If unsafe, a second line must include a comma-separated list of violated categories.<|eot_id|><|start_header_id|>assistant<|end_header_id|>" }}
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_12/special_tokens_map.json b/fed-learning/non-dp/tuned-llama-guard3-round_12/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..92d7fab6729d723f45689803b99006fec945cc47
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_12/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b1835caa5b4d70acaa210fa222b0036f1882f9525c4660fd4810fb3e1e40ff8
+size 325
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_12/tokenizer.json b/fed-learning/non-dp/tuned-llama-guard3-round_12/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af2d22fff23798bea98b1730ae7cdacaee0b087a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_12/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ade1dac458f86f9bea8bf35b713f14e1bbed24228429534038e9f7e54ea3e8b6
+size 17208712
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_12/tokenizer_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_12/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..32408241ddf390e43ec24f8cac8608a7c66f9d1f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_12/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc2a3cb7e7ae6678f28b4a1faee31a00e0f75c475a0fd6af06a100c7f0cde0e5
+size 50550
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_13/README.md b/fed-learning/non-dp/tuned-llama-guard3-round_13/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e544b712adee647f829c2ca8c4571ab05915547a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_13/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Llama-Guard-3-8B
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_13/adapter_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_13/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..329293a4ef5d7c22bff7ad8a59dc4b7622ffc430
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_13/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a1ab765d5bfa0b7d3c35ff748fb8f1813a8150d13a764a19cb6acd2f61d77de
+size 810
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_13/adapter_model.safetensors b/fed-learning/non-dp/tuned-llama-guard3-round_13/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..553bb9d9145f7e77455bb0b76f1f25e45bf228da
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_13/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f55528c840e8f0037a342825e95c41b3c7935402adf91f7372f8e7c9b1885dc5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_13/chat_template.jinja b/fed-learning/non-dp/tuned-llama-guard3-round_13/chat_template.jinja
new file mode 100644
index 0000000000000000000000000000000000000000..8e734bc48af8bcfeecc2d0d1736179ed81008826
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_13/chat_template.jinja
@@ -0,0 +1,30 @@
+{% if messages|length % 2 == 0 %}{% set role = 'Agent' %}{% else %}{% set role = 'User' %}{% endif %}{{ " <|begin_of_text|><|start_header_id|>user<|end_header_id|>
+
+Task: Check if there is unsafe content in '" + role + "' messages in conversations according our safety policy with the below categories.
+
+<BEGIN UNSAFE CONTENT CATEGORIES>
+S1: Violent Crimes.
+S2: Non-Violent Crimes.
+S3: Sex Crimes.
+S4: Child Exploitation.
+S5: Defamation.
+S6: Specialized Advice.
+S7: Privacy.
+S8: Intellectual Property.
+S9: Indiscriminate Weapons.
+S10: Hate.
+S11: Self-Harm.
+S12: Sexual Content.
+S13: Elections.
+S14: Code Interpreter Abuse.
+<END UNSAFE CONTENT CATEGORIES>
+
+<BEGIN CONVERSATION>
+
+" }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{% set role = 'User' %}{% elif message['role'] == 'assistant' %}{% set role = 'Agent' %}{% endif %}{{ role + ': ' + content.strip() + '
+
+' }}{% endfor %}{{ "<END CONVERSATION>
+
+Provide your safety assessment for ONLY THE LAST " + role + " message in the above conversation:
+ - First line must read 'safe' or 'unsafe'.
+ - If unsafe, a second line must include a comma-separated list of violated categories.<|eot_id|><|start_header_id|>assistant<|end_header_id|>" }}
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_13/special_tokens_map.json b/fed-learning/non-dp/tuned-llama-guard3-round_13/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..92d7fab6729d723f45689803b99006fec945cc47
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_13/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b1835caa5b4d70acaa210fa222b0036f1882f9525c4660fd4810fb3e1e40ff8
+size 325
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_13/tokenizer.json b/fed-learning/non-dp/tuned-llama-guard3-round_13/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af2d22fff23798bea98b1730ae7cdacaee0b087a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_13/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ade1dac458f86f9bea8bf35b713f14e1bbed24228429534038e9f7e54ea3e8b6
+size 17208712
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_13/tokenizer_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_13/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..32408241ddf390e43ec24f8cac8608a7c66f9d1f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_13/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc2a3cb7e7ae6678f28b4a1faee31a00e0f75c475a0fd6af06a100c7f0cde0e5
+size 50550
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_14/README.md b/fed-learning/non-dp/tuned-llama-guard3-round_14/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e544b712adee647f829c2ca8c4571ab05915547a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_14/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Llama-Guard-3-8B
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_14/adapter_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_14/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..329293a4ef5d7c22bff7ad8a59dc4b7622ffc430
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_14/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a1ab765d5bfa0b7d3c35ff748fb8f1813a8150d13a764a19cb6acd2f61d77de
+size 810
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_14/adapter_model.safetensors b/fed-learning/non-dp/tuned-llama-guard3-round_14/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..553bb9d9145f7e77455bb0b76f1f25e45bf228da
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_14/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f55528c840e8f0037a342825e95c41b3c7935402adf91f7372f8e7c9b1885dc5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_14/chat_template.jinja b/fed-learning/non-dp/tuned-llama-guard3-round_14/chat_template.jinja
new file mode 100644
index 0000000000000000000000000000000000000000..8e734bc48af8bcfeecc2d0d1736179ed81008826
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_14/chat_template.jinja
@@ -0,0 +1,30 @@
+{% if messages|length % 2 == 0 %}{% set role = 'Agent' %}{% else %}{% set role = 'User' %}{% endif %}{{ " <|begin_of_text|><|start_header_id|>user<|end_header_id|>
+
+Task: Check if there is unsafe content in '" + role + "' messages in conversations according our safety policy with the below categories.
+
+<BEGIN UNSAFE CONTENT CATEGORIES>
+S1: Violent Crimes.
+S2: Non-Violent Crimes.
+S3: Sex Crimes.
+S4: Child Exploitation.
+S5: Defamation.
+S6: Specialized Advice.
+S7: Privacy.
+S8: Intellectual Property.
+S9: Indiscriminate Weapons.
+S10: Hate.
+S11: Self-Harm.
+S12: Sexual Content.
+S13: Elections.
+S14: Code Interpreter Abuse.
+<END UNSAFE CONTENT CATEGORIES>
+
+<BEGIN CONVERSATION>
+
+" }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{% set role = 'User' %}{% elif message['role'] == 'assistant' %}{% set role = 'Agent' %}{% endif %}{{ role + ': ' + content.strip() + '
+
+' }}{% endfor %}{{ "<END CONVERSATION>
+
+Provide your safety assessment for ONLY THE LAST " + role + " message in the above conversation:
+ - First line must read 'safe' or 'unsafe'.
+ - If unsafe, a second line must include a comma-separated list of violated categories.<|eot_id|><|start_header_id|>assistant<|end_header_id|>" }}
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_14/special_tokens_map.json b/fed-learning/non-dp/tuned-llama-guard3-round_14/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..92d7fab6729d723f45689803b99006fec945cc47
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_14/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b1835caa5b4d70acaa210fa222b0036f1882f9525c4660fd4810fb3e1e40ff8
+size 325
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_14/tokenizer.json b/fed-learning/non-dp/tuned-llama-guard3-round_14/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af2d22fff23798bea98b1730ae7cdacaee0b087a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_14/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ade1dac458f86f9bea8bf35b713f14e1bbed24228429534038e9f7e54ea3e8b6
+size 17208712
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_14/tokenizer_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_14/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..32408241ddf390e43ec24f8cac8608a7c66f9d1f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_14/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc2a3cb7e7ae6678f28b4a1faee31a00e0f75c475a0fd6af06a100c7f0cde0e5
+size 50550
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_15/README.md b/fed-learning/non-dp/tuned-llama-guard3-round_15/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e544b712adee647f829c2ca8c4571ab05915547a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_15/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Llama-Guard-3-8B
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_15/adapter_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_15/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..329293a4ef5d7c22bff7ad8a59dc4b7622ffc430
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_15/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a1ab765d5bfa0b7d3c35ff748fb8f1813a8150d13a764a19cb6acd2f61d77de
+size 810
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_15/adapter_model.safetensors b/fed-learning/non-dp/tuned-llama-guard3-round_15/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..553bb9d9145f7e77455bb0b76f1f25e45bf228da
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_15/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f55528c840e8f0037a342825e95c41b3c7935402adf91f7372f8e7c9b1885dc5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_15/chat_template.jinja b/fed-learning/non-dp/tuned-llama-guard3-round_15/chat_template.jinja
new file mode 100644
index 0000000000000000000000000000000000000000..8e734bc48af8bcfeecc2d0d1736179ed81008826
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_15/chat_template.jinja
@@ -0,0 +1,30 @@
+{% if messages|length % 2 == 0 %}{% set role = 'Agent' %}{% else %}{% set role = 'User' %}{% endif %}{{ " <|begin_of_text|><|start_header_id|>user<|end_header_id|>
+
+Task: Check if there is unsafe content in '" + role + "' messages in conversations according our safety policy with the below categories.
+
+<BEGIN UNSAFE CONTENT CATEGORIES>
+S1: Violent Crimes.
+S2: Non-Violent Crimes.
+S3: Sex Crimes.
+S4: Child Exploitation.
+S5: Defamation.
+S6: Specialized Advice.
+S7: Privacy.
+S8: Intellectual Property.
+S9: Indiscriminate Weapons.
+S10: Hate.
+S11: Self-Harm.
+S12: Sexual Content.
+S13: Elections.
+S14: Code Interpreter Abuse.
+<END UNSAFE CONTENT CATEGORIES>
+
+<BEGIN CONVERSATION>
+
+" }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{% set role = 'User' %}{% elif message['role'] == 'assistant' %}{% set role = 'Agent' %}{% endif %}{{ role + ': ' + content.strip() + '
+
+' }}{% endfor %}{{ "<END CONVERSATION>
+
+Provide your safety assessment for ONLY THE LAST " + role + " message in the above conversation:
+ - First line must read 'safe' or 'unsafe'.
+ - If unsafe, a second line must include a comma-separated list of violated categories.<|eot_id|><|start_header_id|>assistant<|end_header_id|>" }}
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_15/special_tokens_map.json b/fed-learning/non-dp/tuned-llama-guard3-round_15/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..92d7fab6729d723f45689803b99006fec945cc47
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_15/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b1835caa5b4d70acaa210fa222b0036f1882f9525c4660fd4810fb3e1e40ff8
+size 325
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_15/tokenizer.json b/fed-learning/non-dp/tuned-llama-guard3-round_15/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af2d22fff23798bea98b1730ae7cdacaee0b087a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_15/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ade1dac458f86f9bea8bf35b713f14e1bbed24228429534038e9f7e54ea3e8b6
+size 17208712
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_15/tokenizer_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_15/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..32408241ddf390e43ec24f8cac8608a7c66f9d1f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_15/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc2a3cb7e7ae6678f28b4a1faee31a00e0f75c475a0fd6af06a100c7f0cde0e5
+size 50550
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_16/README.md b/fed-learning/non-dp/tuned-llama-guard3-round_16/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e544b712adee647f829c2ca8c4571ab05915547a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_16/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Llama-Guard-3-8B
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_16/adapter_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_16/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..329293a4ef5d7c22bff7ad8a59dc4b7622ffc430
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_16/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a1ab765d5bfa0b7d3c35ff748fb8f1813a8150d13a764a19cb6acd2f61d77de
+size 810
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_16/adapter_model.safetensors b/fed-learning/non-dp/tuned-llama-guard3-round_16/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..553bb9d9145f7e77455bb0b76f1f25e45bf228da
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_16/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f55528c840e8f0037a342825e95c41b3c7935402adf91f7372f8e7c9b1885dc5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_16/chat_template.jinja b/fed-learning/non-dp/tuned-llama-guard3-round_16/chat_template.jinja
new file mode 100644
index 0000000000000000000000000000000000000000..8e734bc48af8bcfeecc2d0d1736179ed81008826
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_16/chat_template.jinja
@@ -0,0 +1,30 @@
+{% if messages|length % 2 == 0 %}{% set role = 'Agent' %}{% else %}{% set role = 'User' %}{% endif %}{{ " <|begin_of_text|><|start_header_id|>user<|end_header_id|>
+
+Task: Check if there is unsafe content in '" + role + "' messages in conversations according our safety policy with the below categories.
+
+<BEGIN UNSAFE CONTENT CATEGORIES>
+S1: Violent Crimes.
+S2: Non-Violent Crimes.
+S3: Sex Crimes.
+S4: Child Exploitation.
+S5: Defamation.
+S6: Specialized Advice.
+S7: Privacy.
+S8: Intellectual Property.
+S9: Indiscriminate Weapons.
+S10: Hate.
+S11: Self-Harm.
+S12: Sexual Content.
+S13: Elections.
+S14: Code Interpreter Abuse.
+<END UNSAFE CONTENT CATEGORIES>
+
+<BEGIN CONVERSATION>
+
+" }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{% set role = 'User' %}{% elif message['role'] == 'assistant' %}{% set role = 'Agent' %}{% endif %}{{ role + ': ' + content.strip() + '
+
+' }}{% endfor %}{{ "<END CONVERSATION>
+
+Provide your safety assessment for ONLY THE LAST " + role + " message in the above conversation:
+ - First line must read 'safe' or 'unsafe'.
+ - If unsafe, a second line must include a comma-separated list of violated categories.<|eot_id|><|start_header_id|>assistant<|end_header_id|>" }}
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_16/special_tokens_map.json b/fed-learning/non-dp/tuned-llama-guard3-round_16/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..92d7fab6729d723f45689803b99006fec945cc47
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_16/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b1835caa5b4d70acaa210fa222b0036f1882f9525c4660fd4810fb3e1e40ff8
+size 325
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_16/tokenizer.json b/fed-learning/non-dp/tuned-llama-guard3-round_16/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af2d22fff23798bea98b1730ae7cdacaee0b087a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_16/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ade1dac458f86f9bea8bf35b713f14e1bbed24228429534038e9f7e54ea3e8b6
+size 17208712
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_16/tokenizer_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_16/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..32408241ddf390e43ec24f8cac8608a7c66f9d1f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_16/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc2a3cb7e7ae6678f28b4a1faee31a00e0f75c475a0fd6af06a100c7f0cde0e5
+size 50550
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_17/README.md b/fed-learning/non-dp/tuned-llama-guard3-round_17/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e544b712adee647f829c2ca8c4571ab05915547a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_17/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Llama-Guard-3-8B
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_17/adapter_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_17/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..329293a4ef5d7c22bff7ad8a59dc4b7622ffc430
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_17/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a1ab765d5bfa0b7d3c35ff748fb8f1813a8150d13a764a19cb6acd2f61d77de
+size 810
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_17/adapter_model.safetensors b/fed-learning/non-dp/tuned-llama-guard3-round_17/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..553bb9d9145f7e77455bb0b76f1f25e45bf228da
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_17/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f55528c840e8f0037a342825e95c41b3c7935402adf91f7372f8e7c9b1885dc5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_17/chat_template.jinja b/fed-learning/non-dp/tuned-llama-guard3-round_17/chat_template.jinja
new file mode 100644
index 0000000000000000000000000000000000000000..8e734bc48af8bcfeecc2d0d1736179ed81008826
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_17/chat_template.jinja
@@ -0,0 +1,30 @@
+{% if messages|length % 2 == 0 %}{% set role = 'Agent' %}{% else %}{% set role = 'User' %}{% endif %}{{ " <|begin_of_text|><|start_header_id|>user<|end_header_id|>
+
+Task: Check if there is unsafe content in '" + role + "' messages in conversations according our safety policy with the below categories.
+
+<BEGIN UNSAFE CONTENT CATEGORIES>
+S1: Violent Crimes.
+S2: Non-Violent Crimes.
+S3: Sex Crimes.
+S4: Child Exploitation.
+S5: Defamation.
+S6: Specialized Advice.
+S7: Privacy.
+S8: Intellectual Property.
+S9: Indiscriminate Weapons.
+S10: Hate.
+S11: Self-Harm.
+S12: Sexual Content.
+S13: Elections.
+S14: Code Interpreter Abuse.
+<END UNSAFE CONTENT CATEGORIES>
+
+<BEGIN CONVERSATION>
+
+" }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{% set role = 'User' %}{% elif message['role'] == 'assistant' %}{% set role = 'Agent' %}{% endif %}{{ role + ': ' + content.strip() + '
+
+' }}{% endfor %}{{ "<END CONVERSATION>
+
+Provide your safety assessment for ONLY THE LAST " + role + " message in the above conversation:
+ - First line must read 'safe' or 'unsafe'.
+ - If unsafe, a second line must include a comma-separated list of violated categories.<|eot_id|><|start_header_id|>assistant<|end_header_id|>" }}
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_17/special_tokens_map.json b/fed-learning/non-dp/tuned-llama-guard3-round_17/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..92d7fab6729d723f45689803b99006fec945cc47
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_17/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b1835caa5b4d70acaa210fa222b0036f1882f9525c4660fd4810fb3e1e40ff8
+size 325
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_17/tokenizer.json b/fed-learning/non-dp/tuned-llama-guard3-round_17/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af2d22fff23798bea98b1730ae7cdacaee0b087a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_17/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ade1dac458f86f9bea8bf35b713f14e1bbed24228429534038e9f7e54ea3e8b6
+size 17208712
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_17/tokenizer_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_17/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..32408241ddf390e43ec24f8cac8608a7c66f9d1f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_17/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc2a3cb7e7ae6678f28b4a1faee31a00e0f75c475a0fd6af06a100c7f0cde0e5
+size 50550
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_18/README.md b/fed-learning/non-dp/tuned-llama-guard3-round_18/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e544b712adee647f829c2ca8c4571ab05915547a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_18/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Llama-Guard-3-8B
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_18/adapter_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_18/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..329293a4ef5d7c22bff7ad8a59dc4b7622ffc430
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_18/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a1ab765d5bfa0b7d3c35ff748fb8f1813a8150d13a764a19cb6acd2f61d77de
+size 810
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_18/adapter_model.safetensors b/fed-learning/non-dp/tuned-llama-guard3-round_18/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..553bb9d9145f7e77455bb0b76f1f25e45bf228da
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_18/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f55528c840e8f0037a342825e95c41b3c7935402adf91f7372f8e7c9b1885dc5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_18/chat_template.jinja b/fed-learning/non-dp/tuned-llama-guard3-round_18/chat_template.jinja
new file mode 100644
index 0000000000000000000000000000000000000000..8e734bc48af8bcfeecc2d0d1736179ed81008826
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_18/chat_template.jinja
@@ -0,0 +1,30 @@
+{% if messages|length % 2 == 0 %}{% set role = 'Agent' %}{% else %}{% set role = 'User' %}{% endif %}{{ " <|begin_of_text|><|start_header_id|>user<|end_header_id|>
+
+Task: Check if there is unsafe content in '" + role + "' messages in conversations according our safety policy with the below categories.
+
+<BEGIN UNSAFE CONTENT CATEGORIES>
+S1: Violent Crimes.
+S2: Non-Violent Crimes.
+S3: Sex Crimes.
+S4: Child Exploitation.
+S5: Defamation.
+S6: Specialized Advice.
+S7: Privacy.
+S8: Intellectual Property.
+S9: Indiscriminate Weapons.
+S10: Hate.
+S11: Self-Harm.
+S12: Sexual Content.
+S13: Elections.
+S14: Code Interpreter Abuse.
+<END UNSAFE CONTENT CATEGORIES>
+
+<BEGIN CONVERSATION>
+
+" }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{% set role = 'User' %}{% elif message['role'] == 'assistant' %}{% set role = 'Agent' %}{% endif %}{{ role + ': ' + content.strip() + '
+
+' }}{% endfor %}{{ "<END CONVERSATION>
+
+Provide your safety assessment for ONLY THE LAST " + role + " message in the above conversation:
+ - First line must read 'safe' or 'unsafe'.
+ - If unsafe, a second line must include a comma-separated list of violated categories.<|eot_id|><|start_header_id|>assistant<|end_header_id|>" }}
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_18/special_tokens_map.json b/fed-learning/non-dp/tuned-llama-guard3-round_18/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..92d7fab6729d723f45689803b99006fec945cc47
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_18/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b1835caa5b4d70acaa210fa222b0036f1882f9525c4660fd4810fb3e1e40ff8
+size 325
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_18/tokenizer.json b/fed-learning/non-dp/tuned-llama-guard3-round_18/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af2d22fff23798bea98b1730ae7cdacaee0b087a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_18/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ade1dac458f86f9bea8bf35b713f14e1bbed24228429534038e9f7e54ea3e8b6
+size 17208712
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_18/tokenizer_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_18/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..32408241ddf390e43ec24f8cac8608a7c66f9d1f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_18/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc2a3cb7e7ae6678f28b4a1faee31a00e0f75c475a0fd6af06a100c7f0cde0e5
+size 50550
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_19/README.md b/fed-learning/non-dp/tuned-llama-guard3-round_19/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e544b712adee647f829c2ca8c4571ab05915547a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_19/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Llama-Guard-3-8B
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_19/adapter_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_19/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..329293a4ef5d7c22bff7ad8a59dc4b7622ffc430
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_19/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a1ab765d5bfa0b7d3c35ff748fb8f1813a8150d13a764a19cb6acd2f61d77de
+size 810
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_19/adapter_model.safetensors b/fed-learning/non-dp/tuned-llama-guard3-round_19/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..553bb9d9145f7e77455bb0b76f1f25e45bf228da
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_19/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f55528c840e8f0037a342825e95c41b3c7935402adf91f7372f8e7c9b1885dc5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_19/chat_template.jinja b/fed-learning/non-dp/tuned-llama-guard3-round_19/chat_template.jinja
new file mode 100644
index 0000000000000000000000000000000000000000..8e734bc48af8bcfeecc2d0d1736179ed81008826
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_19/chat_template.jinja
@@ -0,0 +1,30 @@
+{% if messages|length % 2 == 0 %}{% set role = 'Agent' %}{% else %}{% set role = 'User' %}{% endif %}{{ " <|begin_of_text|><|start_header_id|>user<|end_header_id|>
+
+Task: Check if there is unsafe content in '" + role + "' messages in conversations according our safety policy with the below categories.
+
+<BEGIN UNSAFE CONTENT CATEGORIES>
+S1: Violent Crimes.
+S2: Non-Violent Crimes.
+S3: Sex Crimes.
+S4: Child Exploitation.
+S5: Defamation.
+S6: Specialized Advice.
+S7: Privacy.
+S8: Intellectual Property.
+S9: Indiscriminate Weapons.
+S10: Hate.
+S11: Self-Harm.
+S12: Sexual Content.
+S13: Elections.
+S14: Code Interpreter Abuse.
+<END UNSAFE CONTENT CATEGORIES>
+
+<BEGIN CONVERSATION>
+
+" }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{% set role = 'User' %}{% elif message['role'] == 'assistant' %}{% set role = 'Agent' %}{% endif %}{{ role + ': ' + content.strip() + '
+
+' }}{% endfor %}{{ "<END CONVERSATION>
+
+Provide your safety assessment for ONLY THE LAST " + role + " message in the above conversation:
+ - First line must read 'safe' or 'unsafe'.
+ - If unsafe, a second line must include a comma-separated list of violated categories.<|eot_id|><|start_header_id|>assistant<|end_header_id|>" }}
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_19/special_tokens_map.json b/fed-learning/non-dp/tuned-llama-guard3-round_19/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..92d7fab6729d723f45689803b99006fec945cc47
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_19/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b1835caa5b4d70acaa210fa222b0036f1882f9525c4660fd4810fb3e1e40ff8
+size 325
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_19/tokenizer.json b/fed-learning/non-dp/tuned-llama-guard3-round_19/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af2d22fff23798bea98b1730ae7cdacaee0b087a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_19/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ade1dac458f86f9bea8bf35b713f14e1bbed24228429534038e9f7e54ea3e8b6
+size 17208712
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_19/tokenizer_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_19/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..32408241ddf390e43ec24f8cac8608a7c66f9d1f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_19/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc2a3cb7e7ae6678f28b4a1faee31a00e0f75c475a0fd6af06a100c7f0cde0e5
+size 50550
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_2/README.md b/fed-learning/non-dp/tuned-llama-guard3-round_2/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e544b712adee647f829c2ca8c4571ab05915547a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_2/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Llama-Guard-3-8B
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_2/adapter_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_2/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..329293a4ef5d7c22bff7ad8a59dc4b7622ffc430
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_2/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a1ab765d5bfa0b7d3c35ff748fb8f1813a8150d13a764a19cb6acd2f61d77de
+size 810
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_2/adapter_model.safetensors b/fed-learning/non-dp/tuned-llama-guard3-round_2/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..e1aeb0a339488c03ac34117abcd8ae46ab045712
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_2/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b2afccd1b29eecc361410371260d17b9baba64d9777e72b66957da1e6ae8f06
+size 27297032
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_2/chat_template.jinja b/fed-learning/non-dp/tuned-llama-guard3-round_2/chat_template.jinja
new file mode 100644
index 0000000000000000000000000000000000000000..8e734bc48af8bcfeecc2d0d1736179ed81008826
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_2/chat_template.jinja
@@ -0,0 +1,30 @@
+{% if messages|length % 2 == 0 %}{% set role = 'Agent' %}{% else %}{% set role = 'User' %}{% endif %}{{ " <|begin_of_text|><|start_header_id|>user<|end_header_id|>
+
+Task: Check if there is unsafe content in '" + role + "' messages in conversations according our safety policy with the below categories.
+
+<BEGIN UNSAFE CONTENT CATEGORIES>
+S1: Violent Crimes.
+S2: Non-Violent Crimes.
+S3: Sex Crimes.
+S4: Child Exploitation.
+S5: Defamation.
+S6: Specialized Advice.
+S7: Privacy.
+S8: Intellectual Property.
+S9: Indiscriminate Weapons.
+S10: Hate.
+S11: Self-Harm.
+S12: Sexual Content.
+S13: Elections.
+S14: Code Interpreter Abuse.
+<END UNSAFE CONTENT CATEGORIES>
+
+<BEGIN CONVERSATION>
+
+" }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{% set role = 'User' %}{% elif message['role'] == 'assistant' %}{% set role = 'Agent' %}{% endif %}{{ role + ': ' + content.strip() + '
+
+' }}{% endfor %}{{ "<END CONVERSATION>
+
+Provide your safety assessment for ONLY THE LAST " + role + " message in the above conversation:
+ - First line must read 'safe' or 'unsafe'.
+ - If unsafe, a second line must include a comma-separated list of violated categories.<|eot_id|><|start_header_id|>assistant<|end_header_id|>" }}
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_2/special_tokens_map.json b/fed-learning/non-dp/tuned-llama-guard3-round_2/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..92d7fab6729d723f45689803b99006fec945cc47
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_2/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b1835caa5b4d70acaa210fa222b0036f1882f9525c4660fd4810fb3e1e40ff8
+size 325
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_2/tokenizer.json b/fed-learning/non-dp/tuned-llama-guard3-round_2/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af2d22fff23798bea98b1730ae7cdacaee0b087a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_2/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ade1dac458f86f9bea8bf35b713f14e1bbed24228429534038e9f7e54ea3e8b6
+size 17208712
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_2/tokenizer_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_2/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..32408241ddf390e43ec24f8cac8608a7c66f9d1f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_2/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc2a3cb7e7ae6678f28b4a1faee31a00e0f75c475a0fd6af06a100c7f0cde0e5
+size 50550
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_20/README.md b/fed-learning/non-dp/tuned-llama-guard3-round_20/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e544b712adee647f829c2ca8c4571ab05915547a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_20/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Llama-Guard-3-8B
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_20/adapter_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_20/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..329293a4ef5d7c22bff7ad8a59dc4b7622ffc430
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_20/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a1ab765d5bfa0b7d3c35ff748fb8f1813a8150d13a764a19cb6acd2f61d77de
+size 810
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_20/adapter_model.safetensors b/fed-learning/non-dp/tuned-llama-guard3-round_20/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..553bb9d9145f7e77455bb0b76f1f25e45bf228da
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_20/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f55528c840e8f0037a342825e95c41b3c7935402adf91f7372f8e7c9b1885dc5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_20/chat_template.jinja b/fed-learning/non-dp/tuned-llama-guard3-round_20/chat_template.jinja
new file mode 100644
index 0000000000000000000000000000000000000000..8e734bc48af8bcfeecc2d0d1736179ed81008826
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_20/chat_template.jinja
@@ -0,0 +1,30 @@
+{% if messages|length % 2 == 0 %}{% set role = 'Agent' %}{% else %}{% set role = 'User' %}{% endif %}{{ " <|begin_of_text|><|start_header_id|>user<|end_header_id|>
+
+Task: Check if there is unsafe content in '" + role + "' messages in conversations according our safety policy with the below categories.
+
+<BEGIN UNSAFE CONTENT CATEGORIES>
+S1: Violent Crimes.
+S2: Non-Violent Crimes.
+S3: Sex Crimes.
+S4: Child Exploitation.
+S5: Defamation.
+S6: Specialized Advice.
+S7: Privacy.
+S8: Intellectual Property.
+S9: Indiscriminate Weapons.
+S10: Hate.
+S11: Self-Harm.
+S12: Sexual Content.
+S13: Elections.
+S14: Code Interpreter Abuse.
+<END UNSAFE CONTENT CATEGORIES>
+
+<BEGIN CONVERSATION>
+
+" }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{% set role = 'User' %}{% elif message['role'] == 'assistant' %}{% set role = 'Agent' %}{% endif %}{{ role + ': ' + content.strip() + '
+
+' }}{% endfor %}{{ "<END CONVERSATION>
+
+Provide your safety assessment for ONLY THE LAST " + role + " message in the above conversation:
+ - First line must read 'safe' or 'unsafe'.
+ - If unsafe, a second line must include a comma-separated list of violated categories.<|eot_id|><|start_header_id|>assistant<|end_header_id|>" }}
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_20/special_tokens_map.json b/fed-learning/non-dp/tuned-llama-guard3-round_20/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..92d7fab6729d723f45689803b99006fec945cc47
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_20/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b1835caa5b4d70acaa210fa222b0036f1882f9525c4660fd4810fb3e1e40ff8
+size 325
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_20/tokenizer.json b/fed-learning/non-dp/tuned-llama-guard3-round_20/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af2d22fff23798bea98b1730ae7cdacaee0b087a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_20/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ade1dac458f86f9bea8bf35b713f14e1bbed24228429534038e9f7e54ea3e8b6
+size 17208712
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_20/tokenizer_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_20/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..32408241ddf390e43ec24f8cac8608a7c66f9d1f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_20/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc2a3cb7e7ae6678f28b4a1faee31a00e0f75c475a0fd6af06a100c7f0cde0e5
+size 50550
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_21/README.md b/fed-learning/non-dp/tuned-llama-guard3-round_21/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e544b712adee647f829c2ca8c4571ab05915547a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_21/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Llama-Guard-3-8B
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_21/adapter_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_21/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..329293a4ef5d7c22bff7ad8a59dc4b7622ffc430
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_21/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a1ab765d5bfa0b7d3c35ff748fb8f1813a8150d13a764a19cb6acd2f61d77de
+size 810
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_21/adapter_model.safetensors b/fed-learning/non-dp/tuned-llama-guard3-round_21/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..553bb9d9145f7e77455bb0b76f1f25e45bf228da
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_21/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f55528c840e8f0037a342825e95c41b3c7935402adf91f7372f8e7c9b1885dc5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_21/chat_template.jinja b/fed-learning/non-dp/tuned-llama-guard3-round_21/chat_template.jinja
new file mode 100644
index 0000000000000000000000000000000000000000..8e734bc48af8bcfeecc2d0d1736179ed81008826
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_21/chat_template.jinja
@@ -0,0 +1,30 @@
+{% if messages|length % 2 == 0 %}{% set role = 'Agent' %}{% else %}{% set role = 'User' %}{% endif %}{{ " <|begin_of_text|><|start_header_id|>user<|end_header_id|>
+
+Task: Check if there is unsafe content in '" + role + "' messages in conversations according our safety policy with the below categories.
+
+<BEGIN UNSAFE CONTENT CATEGORIES>
+S1: Violent Crimes.
+S2: Non-Violent Crimes.
+S3: Sex Crimes.
+S4: Child Exploitation.
+S5: Defamation.
+S6: Specialized Advice.
+S7: Privacy.
+S8: Intellectual Property.
+S9: Indiscriminate Weapons.
+S10: Hate.
+S11: Self-Harm.
+S12: Sexual Content.
+S13: Elections.
+S14: Code Interpreter Abuse.
+<END UNSAFE CONTENT CATEGORIES>
+
+<BEGIN CONVERSATION>
+
+" }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{% set role = 'User' %}{% elif message['role'] == 'assistant' %}{% set role = 'Agent' %}{% endif %}{{ role + ': ' + content.strip() + '
+
+' }}{% endfor %}{{ "<END CONVERSATION>
+
+Provide your safety assessment for ONLY THE LAST " + role + " message in the above conversation:
+ - First line must read 'safe' or 'unsafe'.
+ - If unsafe, a second line must include a comma-separated list of violated categories.<|eot_id|><|start_header_id|>assistant<|end_header_id|>" }}
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_21/special_tokens_map.json b/fed-learning/non-dp/tuned-llama-guard3-round_21/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..92d7fab6729d723f45689803b99006fec945cc47
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_21/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b1835caa5b4d70acaa210fa222b0036f1882f9525c4660fd4810fb3e1e40ff8
+size 325
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_21/tokenizer.json b/fed-learning/non-dp/tuned-llama-guard3-round_21/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af2d22fff23798bea98b1730ae7cdacaee0b087a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_21/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ade1dac458f86f9bea8bf35b713f14e1bbed24228429534038e9f7e54ea3e8b6
+size 17208712
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_21/tokenizer_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_21/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..32408241ddf390e43ec24f8cac8608a7c66f9d1f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_21/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc2a3cb7e7ae6678f28b4a1faee31a00e0f75c475a0fd6af06a100c7f0cde0e5
+size 50550
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_22/README.md b/fed-learning/non-dp/tuned-llama-guard3-round_22/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e544b712adee647f829c2ca8c4571ab05915547a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_22/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Llama-Guard-3-8B
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_22/adapter_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_22/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..329293a4ef5d7c22bff7ad8a59dc4b7622ffc430
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_22/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a1ab765d5bfa0b7d3c35ff748fb8f1813a8150d13a764a19cb6acd2f61d77de
+size 810
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_22/adapter_model.safetensors b/fed-learning/non-dp/tuned-llama-guard3-round_22/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..553bb9d9145f7e77455bb0b76f1f25e45bf228da
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_22/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f55528c840e8f0037a342825e95c41b3c7935402adf91f7372f8e7c9b1885dc5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_22/chat_template.jinja b/fed-learning/non-dp/tuned-llama-guard3-round_22/chat_template.jinja
new file mode 100644
index 0000000000000000000000000000000000000000..8e734bc48af8bcfeecc2d0d1736179ed81008826
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_22/chat_template.jinja
@@ -0,0 +1,30 @@
+{% if messages|length % 2 == 0 %}{% set role = 'Agent' %}{% else %}{% set role = 'User' %}{% endif %}{{ " <|begin_of_text|><|start_header_id|>user<|end_header_id|>
+
+Task: Check if there is unsafe content in '" + role + "' messages in conversations according our safety policy with the below categories.
+
+<BEGIN UNSAFE CONTENT CATEGORIES>
+S1: Violent Crimes.
+S2: Non-Violent Crimes.
+S3: Sex Crimes.
+S4: Child Exploitation.
+S5: Defamation.
+S6: Specialized Advice.
+S7: Privacy.
+S8: Intellectual Property.
+S9: Indiscriminate Weapons.
+S10: Hate.
+S11: Self-Harm.
+S12: Sexual Content.
+S13: Elections.
+S14: Code Interpreter Abuse.
+<END UNSAFE CONTENT CATEGORIES>
+
+<BEGIN CONVERSATION>
+
+" }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{% set role = 'User' %}{% elif message['role'] == 'assistant' %}{% set role = 'Agent' %}{% endif %}{{ role + ': ' + content.strip() + '
+
+' }}{% endfor %}{{ "<END CONVERSATION>
+
+Provide your safety assessment for ONLY THE LAST " + role + " message in the above conversation:
+ - First line must read 'safe' or 'unsafe'.
+ - If unsafe, a second line must include a comma-separated list of violated categories.<|eot_id|><|start_header_id|>assistant<|end_header_id|>" }}
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_22/special_tokens_map.json b/fed-learning/non-dp/tuned-llama-guard3-round_22/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..92d7fab6729d723f45689803b99006fec945cc47
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_22/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b1835caa5b4d70acaa210fa222b0036f1882f9525c4660fd4810fb3e1e40ff8
+size 325
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_22/tokenizer.json b/fed-learning/non-dp/tuned-llama-guard3-round_22/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af2d22fff23798bea98b1730ae7cdacaee0b087a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_22/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ade1dac458f86f9bea8bf35b713f14e1bbed24228429534038e9f7e54ea3e8b6
+size 17208712
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_22/tokenizer_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_22/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..32408241ddf390e43ec24f8cac8608a7c66f9d1f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_22/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc2a3cb7e7ae6678f28b4a1faee31a00e0f75c475a0fd6af06a100c7f0cde0e5
+size 50550
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_23/README.md b/fed-learning/non-dp/tuned-llama-guard3-round_23/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e544b712adee647f829c2ca8c4571ab05915547a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_23/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Llama-Guard-3-8B
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_23/adapter_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_23/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..329293a4ef5d7c22bff7ad8a59dc4b7622ffc430
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_23/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a1ab765d5bfa0b7d3c35ff748fb8f1813a8150d13a764a19cb6acd2f61d77de
+size 810
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_23/adapter_model.safetensors b/fed-learning/non-dp/tuned-llama-guard3-round_23/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..553bb9d9145f7e77455bb0b76f1f25e45bf228da
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_23/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f55528c840e8f0037a342825e95c41b3c7935402adf91f7372f8e7c9b1885dc5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_23/chat_template.jinja b/fed-learning/non-dp/tuned-llama-guard3-round_23/chat_template.jinja
new file mode 100644
index 0000000000000000000000000000000000000000..8e734bc48af8bcfeecc2d0d1736179ed81008826
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_23/chat_template.jinja
@@ -0,0 +1,30 @@
+{% if messages|length % 2 == 0 %}{% set role = 'Agent' %}{% else %}{% set role = 'User' %}{% endif %}{{ " <|begin_of_text|><|start_header_id|>user<|end_header_id|>
+
+Task: Check if there is unsafe content in '" + role + "' messages in conversations according our safety policy with the below categories.
+
+<BEGIN UNSAFE CONTENT CATEGORIES>
+S1: Violent Crimes.
+S2: Non-Violent Crimes.
+S3: Sex Crimes.
+S4: Child Exploitation.
+S5: Defamation.
+S6: Specialized Advice.
+S7: Privacy.
+S8: Intellectual Property.
+S9: Indiscriminate Weapons.
+S10: Hate.
+S11: Self-Harm.
+S12: Sexual Content.
+S13: Elections.
+S14: Code Interpreter Abuse.
+<END UNSAFE CONTENT CATEGORIES>
+
+<BEGIN CONVERSATION>
+
+" }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{% set role = 'User' %}{% elif message['role'] == 'assistant' %}{% set role = 'Agent' %}{% endif %}{{ role + ': ' + content.strip() + '
+
+' }}{% endfor %}{{ "<END CONVERSATION>
+
+Provide your safety assessment for ONLY THE LAST " + role + " message in the above conversation:
+ - First line must read 'safe' or 'unsafe'.
+ - If unsafe, a second line must include a comma-separated list of violated categories.<|eot_id|><|start_header_id|>assistant<|end_header_id|>" }}
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_23/special_tokens_map.json b/fed-learning/non-dp/tuned-llama-guard3-round_23/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..92d7fab6729d723f45689803b99006fec945cc47
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_23/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b1835caa5b4d70acaa210fa222b0036f1882f9525c4660fd4810fb3e1e40ff8
+size 325
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_23/tokenizer.json b/fed-learning/non-dp/tuned-llama-guard3-round_23/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af2d22fff23798bea98b1730ae7cdacaee0b087a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_23/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ade1dac458f86f9bea8bf35b713f14e1bbed24228429534038e9f7e54ea3e8b6
+size 17208712
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_23/tokenizer_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_23/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..32408241ddf390e43ec24f8cac8608a7c66f9d1f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_23/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc2a3cb7e7ae6678f28b4a1faee31a00e0f75c475a0fd6af06a100c7f0cde0e5
+size 50550
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_24/README.md b/fed-learning/non-dp/tuned-llama-guard3-round_24/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e544b712adee647f829c2ca8c4571ab05915547a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_24/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Llama-Guard-3-8B
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_24/adapter_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_24/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..329293a4ef5d7c22bff7ad8a59dc4b7622ffc430
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_24/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a1ab765d5bfa0b7d3c35ff748fb8f1813a8150d13a764a19cb6acd2f61d77de
+size 810
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_24/adapter_model.safetensors b/fed-learning/non-dp/tuned-llama-guard3-round_24/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..553bb9d9145f7e77455bb0b76f1f25e45bf228da
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_24/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f55528c840e8f0037a342825e95c41b3c7935402adf91f7372f8e7c9b1885dc5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_24/chat_template.jinja b/fed-learning/non-dp/tuned-llama-guard3-round_24/chat_template.jinja
new file mode 100644
index 0000000000000000000000000000000000000000..8e734bc48af8bcfeecc2d0d1736179ed81008826
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_24/chat_template.jinja
@@ -0,0 +1,30 @@
+{% if messages|length % 2 == 0 %}{% set role = 'Agent' %}{% else %}{% set role = 'User' %}{% endif %}{{ " <|begin_of_text|><|start_header_id|>user<|end_header_id|>
+
+Task: Check if there is unsafe content in '" + role + "' messages in conversations according our safety policy with the below categories.
+
+<BEGIN UNSAFE CONTENT CATEGORIES>
+S1: Violent Crimes.
+S2: Non-Violent Crimes.
+S3: Sex Crimes.
+S4: Child Exploitation.
+S5: Defamation.
+S6: Specialized Advice.
+S7: Privacy.
+S8: Intellectual Property.
+S9: Indiscriminate Weapons.
+S10: Hate.
+S11: Self-Harm.
+S12: Sexual Content.
+S13: Elections.
+S14: Code Interpreter Abuse.
+<END UNSAFE CONTENT CATEGORIES>
+
+<BEGIN CONVERSATION>
+
+" }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{% set role = 'User' %}{% elif message['role'] == 'assistant' %}{% set role = 'Agent' %}{% endif %}{{ role + ': ' + content.strip() + '
+
+' }}{% endfor %}{{ "<END CONVERSATION>
+
+Provide your safety assessment for ONLY THE LAST " + role + " message in the above conversation:
+ - First line must read 'safe' or 'unsafe'.
+ - If unsafe, a second line must include a comma-separated list of violated categories.<|eot_id|><|start_header_id|>assistant<|end_header_id|>" }}
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_24/special_tokens_map.json b/fed-learning/non-dp/tuned-llama-guard3-round_24/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..92d7fab6729d723f45689803b99006fec945cc47
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_24/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b1835caa5b4d70acaa210fa222b0036f1882f9525c4660fd4810fb3e1e40ff8
+size 325
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_24/tokenizer.json b/fed-learning/non-dp/tuned-llama-guard3-round_24/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af2d22fff23798bea98b1730ae7cdacaee0b087a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_24/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ade1dac458f86f9bea8bf35b713f14e1bbed24228429534038e9f7e54ea3e8b6
+size 17208712
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_24/tokenizer_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_24/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..32408241ddf390e43ec24f8cac8608a7c66f9d1f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_24/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc2a3cb7e7ae6678f28b4a1faee31a00e0f75c475a0fd6af06a100c7f0cde0e5
+size 50550
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_25/README.md b/fed-learning/non-dp/tuned-llama-guard3-round_25/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e544b712adee647f829c2ca8c4571ab05915547a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_25/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Llama-Guard-3-8B
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_25/adapter_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_25/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..329293a4ef5d7c22bff7ad8a59dc4b7622ffc430
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_25/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a1ab765d5bfa0b7d3c35ff748fb8f1813a8150d13a764a19cb6acd2f61d77de
+size 810
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_25/adapter_model.safetensors b/fed-learning/non-dp/tuned-llama-guard3-round_25/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..553bb9d9145f7e77455bb0b76f1f25e45bf228da
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_25/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f55528c840e8f0037a342825e95c41b3c7935402adf91f7372f8e7c9b1885dc5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_25/chat_template.jinja b/fed-learning/non-dp/tuned-llama-guard3-round_25/chat_template.jinja
new file mode 100644
index 0000000000000000000000000000000000000000..8e734bc48af8bcfeecc2d0d1736179ed81008826
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_25/chat_template.jinja
@@ -0,0 +1,30 @@
+{% if messages|length % 2 == 0 %}{% set role = 'Agent' %}{% else %}{% set role = 'User' %}{% endif %}{{ " <|begin_of_text|><|start_header_id|>user<|end_header_id|>
+
+Task: Check if there is unsafe content in '" + role + "' messages in conversations according our safety policy with the below categories.
+
+<BEGIN UNSAFE CONTENT CATEGORIES>
+S1: Violent Crimes.
+S2: Non-Violent Crimes.
+S3: Sex Crimes.
+S4: Child Exploitation.
+S5: Defamation.
+S6: Specialized Advice.
+S7: Privacy.
+S8: Intellectual Property.
+S9: Indiscriminate Weapons.
+S10: Hate.
+S11: Self-Harm.
+S12: Sexual Content.
+S13: Elections.
+S14: Code Interpreter Abuse.
+<END UNSAFE CONTENT CATEGORIES>
+
+<BEGIN CONVERSATION>
+
+" }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{% set role = 'User' %}{% elif message['role'] == 'assistant' %}{% set role = 'Agent' %}{% endif %}{{ role + ': ' + content.strip() + '
+
+' }}{% endfor %}{{ "<END CONVERSATION>
+
+Provide your safety assessment for ONLY THE LAST " + role + " message in the above conversation:
+ - First line must read 'safe' or 'unsafe'.
+ - If unsafe, a second line must include a comma-separated list of violated categories.<|eot_id|><|start_header_id|>assistant<|end_header_id|>" }}
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_25/special_tokens_map.json b/fed-learning/non-dp/tuned-llama-guard3-round_25/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..92d7fab6729d723f45689803b99006fec945cc47
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_25/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b1835caa5b4d70acaa210fa222b0036f1882f9525c4660fd4810fb3e1e40ff8
+size 325
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_25/tokenizer.json b/fed-learning/non-dp/tuned-llama-guard3-round_25/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af2d22fff23798bea98b1730ae7cdacaee0b087a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_25/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ade1dac458f86f9bea8bf35b713f14e1bbed24228429534038e9f7e54ea3e8b6
+size 17208712
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_25/tokenizer_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_25/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..32408241ddf390e43ec24f8cac8608a7c66f9d1f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_25/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc2a3cb7e7ae6678f28b4a1faee31a00e0f75c475a0fd6af06a100c7f0cde0e5
+size 50550
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_26/README.md b/fed-learning/non-dp/tuned-llama-guard3-round_26/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e544b712adee647f829c2ca8c4571ab05915547a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_26/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Llama-Guard-3-8B
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_26/adapter_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_26/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..329293a4ef5d7c22bff7ad8a59dc4b7622ffc430
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_26/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a1ab765d5bfa0b7d3c35ff748fb8f1813a8150d13a764a19cb6acd2f61d77de
+size 810
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_26/adapter_model.safetensors b/fed-learning/non-dp/tuned-llama-guard3-round_26/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..553bb9d9145f7e77455bb0b76f1f25e45bf228da
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_26/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f55528c840e8f0037a342825e95c41b3c7935402adf91f7372f8e7c9b1885dc5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_26/chat_template.jinja b/fed-learning/non-dp/tuned-llama-guard3-round_26/chat_template.jinja
new file mode 100644
index 0000000000000000000000000000000000000000..8e734bc48af8bcfeecc2d0d1736179ed81008826
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_26/chat_template.jinja
@@ -0,0 +1,30 @@
+{% if messages|length % 2 == 0 %}{% set role = 'Agent' %}{% else %}{% set role = 'User' %}{% endif %}{{ " <|begin_of_text|><|start_header_id|>user<|end_header_id|>
+
+Task: Check if there is unsafe content in '" + role + "' messages in conversations according our safety policy with the below categories.
+
+<BEGIN UNSAFE CONTENT CATEGORIES>
+S1: Violent Crimes.
+S2: Non-Violent Crimes.
+S3: Sex Crimes.
+S4: Child Exploitation.
+S5: Defamation.
+S6: Specialized Advice.
+S7: Privacy.
+S8: Intellectual Property.
+S9: Indiscriminate Weapons.
+S10: Hate.
+S11: Self-Harm.
+S12: Sexual Content.
+S13: Elections.
+S14: Code Interpreter Abuse.
+<END UNSAFE CONTENT CATEGORIES>
+
+<BEGIN CONVERSATION>
+
+" }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{% set role = 'User' %}{% elif message['role'] == 'assistant' %}{% set role = 'Agent' %}{% endif %}{{ role + ': ' + content.strip() + '
+
+' }}{% endfor %}{{ "<END CONVERSATION>
+
+Provide your safety assessment for ONLY THE LAST " + role + " message in the above conversation:
+ - First line must read 'safe' or 'unsafe'.
+ - If unsafe, a second line must include a comma-separated list of violated categories.<|eot_id|><|start_header_id|>assistant<|end_header_id|>" }}
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_26/special_tokens_map.json b/fed-learning/non-dp/tuned-llama-guard3-round_26/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..92d7fab6729d723f45689803b99006fec945cc47
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_26/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b1835caa5b4d70acaa210fa222b0036f1882f9525c4660fd4810fb3e1e40ff8
+size 325
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_26/tokenizer.json b/fed-learning/non-dp/tuned-llama-guard3-round_26/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af2d22fff23798bea98b1730ae7cdacaee0b087a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_26/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ade1dac458f86f9bea8bf35b713f14e1bbed24228429534038e9f7e54ea3e8b6
+size 17208712
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_26/tokenizer_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_26/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..32408241ddf390e43ec24f8cac8608a7c66f9d1f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_26/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc2a3cb7e7ae6678f28b4a1faee31a00e0f75c475a0fd6af06a100c7f0cde0e5
+size 50550
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_27/README.md b/fed-learning/non-dp/tuned-llama-guard3-round_27/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e544b712adee647f829c2ca8c4571ab05915547a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_27/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Llama-Guard-3-8B
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_27/adapter_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_27/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..329293a4ef5d7c22bff7ad8a59dc4b7622ffc430
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_27/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a1ab765d5bfa0b7d3c35ff748fb8f1813a8150d13a764a19cb6acd2f61d77de
+size 810
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_27/adapter_model.safetensors b/fed-learning/non-dp/tuned-llama-guard3-round_27/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..553bb9d9145f7e77455bb0b76f1f25e45bf228da
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_27/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f55528c840e8f0037a342825e95c41b3c7935402adf91f7372f8e7c9b1885dc5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_27/chat_template.jinja b/fed-learning/non-dp/tuned-llama-guard3-round_27/chat_template.jinja
new file mode 100644
index 0000000000000000000000000000000000000000..8e734bc48af8bcfeecc2d0d1736179ed81008826
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_27/chat_template.jinja
@@ -0,0 +1,30 @@
+{% if messages|length % 2 == 0 %}{% set role = 'Agent' %}{% else %}{% set role = 'User' %}{% endif %}{{ " <|begin_of_text|><|start_header_id|>user<|end_header_id|>
+
+Task: Check if there is unsafe content in '" + role + "' messages in conversations according our safety policy with the below categories.
+
+<BEGIN UNSAFE CONTENT CATEGORIES>
+S1: Violent Crimes.
+S2: Non-Violent Crimes.
+S3: Sex Crimes.
+S4: Child Exploitation.
+S5: Defamation.
+S6: Specialized Advice.
+S7: Privacy.
+S8: Intellectual Property.
+S9: Indiscriminate Weapons.
+S10: Hate.
+S11: Self-Harm.
+S12: Sexual Content.
+S13: Elections.
+S14: Code Interpreter Abuse.
+<END UNSAFE CONTENT CATEGORIES>
+
+<BEGIN CONVERSATION>
+
+" }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{% set role = 'User' %}{% elif message['role'] == 'assistant' %}{% set role = 'Agent' %}{% endif %}{{ role + ': ' + content.strip() + '
+
+' }}{% endfor %}{{ "<END CONVERSATION>
+
+Provide your safety assessment for ONLY THE LAST " + role + " message in the above conversation:
+ - First line must read 'safe' or 'unsafe'.
+ - If unsafe, a second line must include a comma-separated list of violated categories.<|eot_id|><|start_header_id|>assistant<|end_header_id|>" }}
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_27/special_tokens_map.json b/fed-learning/non-dp/tuned-llama-guard3-round_27/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..92d7fab6729d723f45689803b99006fec945cc47
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_27/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b1835caa5b4d70acaa210fa222b0036f1882f9525c4660fd4810fb3e1e40ff8
+size 325
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_27/tokenizer.json b/fed-learning/non-dp/tuned-llama-guard3-round_27/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af2d22fff23798bea98b1730ae7cdacaee0b087a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_27/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ade1dac458f86f9bea8bf35b713f14e1bbed24228429534038e9f7e54ea3e8b6
+size 17208712
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_27/tokenizer_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_27/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..32408241ddf390e43ec24f8cac8608a7c66f9d1f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_27/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc2a3cb7e7ae6678f28b4a1faee31a00e0f75c475a0fd6af06a100c7f0cde0e5
+size 50550
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_28/README.md b/fed-learning/non-dp/tuned-llama-guard3-round_28/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e544b712adee647f829c2ca8c4571ab05915547a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_28/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Llama-Guard-3-8B
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_28/adapter_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_28/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..329293a4ef5d7c22bff7ad8a59dc4b7622ffc430
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_28/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a1ab765d5bfa0b7d3c35ff748fb8f1813a8150d13a764a19cb6acd2f61d77de
+size 810
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_28/adapter_model.safetensors b/fed-learning/non-dp/tuned-llama-guard3-round_28/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..553bb9d9145f7e77455bb0b76f1f25e45bf228da
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_28/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f55528c840e8f0037a342825e95c41b3c7935402adf91f7372f8e7c9b1885dc5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_28/chat_template.jinja b/fed-learning/non-dp/tuned-llama-guard3-round_28/chat_template.jinja
new file mode 100644
index 0000000000000000000000000000000000000000..8e734bc48af8bcfeecc2d0d1736179ed81008826
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_28/chat_template.jinja
@@ -0,0 +1,30 @@
+{% if messages|length % 2 == 0 %}{% set role = 'Agent' %}{% else %}{% set role = 'User' %}{% endif %}{{ " <|begin_of_text|><|start_header_id|>user<|end_header_id|>
+
+Task: Check if there is unsafe content in '" + role + "' messages in conversations according our safety policy with the below categories.
+
+<BEGIN UNSAFE CONTENT CATEGORIES>
+S1: Violent Crimes.
+S2: Non-Violent Crimes.
+S3: Sex Crimes.
+S4: Child Exploitation.
+S5: Defamation.
+S6: Specialized Advice.
+S7: Privacy.
+S8: Intellectual Property.
+S9: Indiscriminate Weapons.
+S10: Hate.
+S11: Self-Harm.
+S12: Sexual Content.
+S13: Elections.
+S14: Code Interpreter Abuse.
+<END UNSAFE CONTENT CATEGORIES>
+
+<BEGIN CONVERSATION>
+
+" }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{% set role = 'User' %}{% elif message['role'] == 'assistant' %}{% set role = 'Agent' %}{% endif %}{{ role + ': ' + content.strip() + '
+
+' }}{% endfor %}{{ "<END CONVERSATION>
+
+Provide your safety assessment for ONLY THE LAST " + role + " message in the above conversation:
+ - First line must read 'safe' or 'unsafe'.
+ - If unsafe, a second line must include a comma-separated list of violated categories.<|eot_id|><|start_header_id|>assistant<|end_header_id|>" }}
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_28/special_tokens_map.json b/fed-learning/non-dp/tuned-llama-guard3-round_28/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..92d7fab6729d723f45689803b99006fec945cc47
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_28/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b1835caa5b4d70acaa210fa222b0036f1882f9525c4660fd4810fb3e1e40ff8
+size 325
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_28/tokenizer.json b/fed-learning/non-dp/tuned-llama-guard3-round_28/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af2d22fff23798bea98b1730ae7cdacaee0b087a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_28/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ade1dac458f86f9bea8bf35b713f14e1bbed24228429534038e9f7e54ea3e8b6
+size 17208712
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_28/tokenizer_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_28/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..32408241ddf390e43ec24f8cac8608a7c66f9d1f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_28/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc2a3cb7e7ae6678f28b4a1faee31a00e0f75c475a0fd6af06a100c7f0cde0e5
+size 50550
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_29/README.md b/fed-learning/non-dp/tuned-llama-guard3-round_29/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e544b712adee647f829c2ca8c4571ab05915547a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_29/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Llama-Guard-3-8B
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_29/adapter_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_29/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..329293a4ef5d7c22bff7ad8a59dc4b7622ffc430
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_29/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a1ab765d5bfa0b7d3c35ff748fb8f1813a8150d13a764a19cb6acd2f61d77de
+size 810
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_29/adapter_model.safetensors b/fed-learning/non-dp/tuned-llama-guard3-round_29/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..553bb9d9145f7e77455bb0b76f1f25e45bf228da
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_29/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f55528c840e8f0037a342825e95c41b3c7935402adf91f7372f8e7c9b1885dc5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_29/chat_template.jinja b/fed-learning/non-dp/tuned-llama-guard3-round_29/chat_template.jinja
new file mode 100644
index 0000000000000000000000000000000000000000..8e734bc48af8bcfeecc2d0d1736179ed81008826
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_29/chat_template.jinja
@@ -0,0 +1,30 @@
+{% if messages|length % 2 == 0 %}{% set role = 'Agent' %}{% else %}{% set role = 'User' %}{% endif %}{{ " <|begin_of_text|><|start_header_id|>user<|end_header_id|>
+
+Task: Check if there is unsafe content in '" + role + "' messages in conversations according our safety policy with the below categories.
+
+<BEGIN UNSAFE CONTENT CATEGORIES>
+S1: Violent Crimes.
+S2: Non-Violent Crimes.
+S3: Sex Crimes.
+S4: Child Exploitation.
+S5: Defamation.
+S6: Specialized Advice.
+S7: Privacy.
+S8: Intellectual Property.
+S9: Indiscriminate Weapons.
+S10: Hate.
+S11: Self-Harm.
+S12: Sexual Content.
+S13: Elections.
+S14: Code Interpreter Abuse.
+<END UNSAFE CONTENT CATEGORIES>
+
+<BEGIN CONVERSATION>
+
+" }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{% set role = 'User' %}{% elif message['role'] == 'assistant' %}{% set role = 'Agent' %}{% endif %}{{ role + ': ' + content.strip() + '
+
+' }}{% endfor %}{{ "<END CONVERSATION>
+
+Provide your safety assessment for ONLY THE LAST " + role + " message in the above conversation:
+ - First line must read 'safe' or 'unsafe'.
+ - If unsafe, a second line must include a comma-separated list of violated categories.<|eot_id|><|start_header_id|>assistant<|end_header_id|>" }}
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_29/special_tokens_map.json b/fed-learning/non-dp/tuned-llama-guard3-round_29/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..92d7fab6729d723f45689803b99006fec945cc47
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_29/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b1835caa5b4d70acaa210fa222b0036f1882f9525c4660fd4810fb3e1e40ff8
+size 325
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_29/tokenizer.json b/fed-learning/non-dp/tuned-llama-guard3-round_29/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af2d22fff23798bea98b1730ae7cdacaee0b087a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_29/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ade1dac458f86f9bea8bf35b713f14e1bbed24228429534038e9f7e54ea3e8b6
+size 17208712
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_29/tokenizer_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_29/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..32408241ddf390e43ec24f8cac8608a7c66f9d1f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_29/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc2a3cb7e7ae6678f28b4a1faee31a00e0f75c475a0fd6af06a100c7f0cde0e5
+size 50550
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_3/README.md b/fed-learning/non-dp/tuned-llama-guard3-round_3/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e544b712adee647f829c2ca8c4571ab05915547a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_3/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Llama-Guard-3-8B
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_3/adapter_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_3/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..329293a4ef5d7c22bff7ad8a59dc4b7622ffc430
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_3/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a1ab765d5bfa0b7d3c35ff748fb8f1813a8150d13a764a19cb6acd2f61d77de
+size 810
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_3/adapter_model.safetensors b/fed-learning/non-dp/tuned-llama-guard3-round_3/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..553bb9d9145f7e77455bb0b76f1f25e45bf228da
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_3/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f55528c840e8f0037a342825e95c41b3c7935402adf91f7372f8e7c9b1885dc5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_3/chat_template.jinja b/fed-learning/non-dp/tuned-llama-guard3-round_3/chat_template.jinja
new file mode 100644
index 0000000000000000000000000000000000000000..8e734bc48af8bcfeecc2d0d1736179ed81008826
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_3/chat_template.jinja
@@ -0,0 +1,30 @@
+{% if messages|length % 2 == 0 %}{% set role = 'Agent' %}{% else %}{% set role = 'User' %}{% endif %}{{ " <|begin_of_text|><|start_header_id|>user<|end_header_id|>
+
+Task: Check if there is unsafe content in '" + role + "' messages in conversations according our safety policy with the below categories.
+
+<BEGIN UNSAFE CONTENT CATEGORIES>
+S1: Violent Crimes.
+S2: Non-Violent Crimes.
+S3: Sex Crimes.
+S4: Child Exploitation.
+S5: Defamation.
+S6: Specialized Advice.
+S7: Privacy.
+S8: Intellectual Property.
+S9: Indiscriminate Weapons.
+S10: Hate.
+S11: Self-Harm.
+S12: Sexual Content.
+S13: Elections.
+S14: Code Interpreter Abuse.
+<END UNSAFE CONTENT CATEGORIES>
+
+<BEGIN CONVERSATION>
+
+" }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{% set role = 'User' %}{% elif message['role'] == 'assistant' %}{% set role = 'Agent' %}{% endif %}{{ role + ': ' + content.strip() + '
+
+' }}{% endfor %}{{ "<END CONVERSATION>
+
+Provide your safety assessment for ONLY THE LAST " + role + " message in the above conversation:
+ - First line must read 'safe' or 'unsafe'.
+ - If unsafe, a second line must include a comma-separated list of violated categories.<|eot_id|><|start_header_id|>assistant<|end_header_id|>" }}
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_3/special_tokens_map.json b/fed-learning/non-dp/tuned-llama-guard3-round_3/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..92d7fab6729d723f45689803b99006fec945cc47
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_3/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b1835caa5b4d70acaa210fa222b0036f1882f9525c4660fd4810fb3e1e40ff8
+size 325
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_3/tokenizer.json b/fed-learning/non-dp/tuned-llama-guard3-round_3/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af2d22fff23798bea98b1730ae7cdacaee0b087a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_3/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ade1dac458f86f9bea8bf35b713f14e1bbed24228429534038e9f7e54ea3e8b6
+size 17208712
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_3/tokenizer_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_3/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..32408241ddf390e43ec24f8cac8608a7c66f9d1f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_3/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc2a3cb7e7ae6678f28b4a1faee31a00e0f75c475a0fd6af06a100c7f0cde0e5
+size 50550
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_30/README.md b/fed-learning/non-dp/tuned-llama-guard3-round_30/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e544b712adee647f829c2ca8c4571ab05915547a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_30/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Llama-Guard-3-8B
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_30/adapter_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_30/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..329293a4ef5d7c22bff7ad8a59dc4b7622ffc430
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_30/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a1ab765d5bfa0b7d3c35ff748fb8f1813a8150d13a764a19cb6acd2f61d77de
+size 810
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_30/adapter_model.safetensors b/fed-learning/non-dp/tuned-llama-guard3-round_30/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..553bb9d9145f7e77455bb0b76f1f25e45bf228da
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_30/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f55528c840e8f0037a342825e95c41b3c7935402adf91f7372f8e7c9b1885dc5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_30/chat_template.jinja b/fed-learning/non-dp/tuned-llama-guard3-round_30/chat_template.jinja
new file mode 100644
index 0000000000000000000000000000000000000000..8e734bc48af8bcfeecc2d0d1736179ed81008826
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_30/chat_template.jinja
@@ -0,0 +1,30 @@
+{% if messages|length % 2 == 0 %}{% set role = 'Agent' %}{% else %}{% set role = 'User' %}{% endif %}{{ " <|begin_of_text|><|start_header_id|>user<|end_header_id|>
+
+Task: Check if there is unsafe content in '" + role + "' messages in conversations according our safety policy with the below categories.
+
+<BEGIN UNSAFE CONTENT CATEGORIES>
+S1: Violent Crimes.
+S2: Non-Violent Crimes.
+S3: Sex Crimes.
+S4: Child Exploitation.
+S5: Defamation.
+S6: Specialized Advice.
+S7: Privacy.
+S8: Intellectual Property.
+S9: Indiscriminate Weapons.
+S10: Hate.
+S11: Self-Harm.
+S12: Sexual Content.
+S13: Elections.
+S14: Code Interpreter Abuse.
+<END UNSAFE CONTENT CATEGORIES>
+
+<BEGIN CONVERSATION>
+
+" }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{% set role = 'User' %}{% elif message['role'] == 'assistant' %}{% set role = 'Agent' %}{% endif %}{{ role + ': ' + content.strip() + '
+
+' }}{% endfor %}{{ "<END CONVERSATION>
+
+Provide your safety assessment for ONLY THE LAST " + role + " message in the above conversation:
+ - First line must read 'safe' or 'unsafe'.
+ - If unsafe, a second line must include a comma-separated list of violated categories.<|eot_id|><|start_header_id|>assistant<|end_header_id|>" }}
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_30/special_tokens_map.json b/fed-learning/non-dp/tuned-llama-guard3-round_30/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..92d7fab6729d723f45689803b99006fec945cc47
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_30/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b1835caa5b4d70acaa210fa222b0036f1882f9525c4660fd4810fb3e1e40ff8
+size 325
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_30/tokenizer.json b/fed-learning/non-dp/tuned-llama-guard3-round_30/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af2d22fff23798bea98b1730ae7cdacaee0b087a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_30/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ade1dac458f86f9bea8bf35b713f14e1bbed24228429534038e9f7e54ea3e8b6
+size 17208712
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_30/tokenizer_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_30/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..32408241ddf390e43ec24f8cac8608a7c66f9d1f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_30/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc2a3cb7e7ae6678f28b4a1faee31a00e0f75c475a0fd6af06a100c7f0cde0e5
+size 50550
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_4/README.md b/fed-learning/non-dp/tuned-llama-guard3-round_4/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e544b712adee647f829c2ca8c4571ab05915547a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_4/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Llama-Guard-3-8B
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_4/adapter_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_4/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..329293a4ef5d7c22bff7ad8a59dc4b7622ffc430
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_4/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a1ab765d5bfa0b7d3c35ff748fb8f1813a8150d13a764a19cb6acd2f61d77de
+size 810
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_4/adapter_model.safetensors b/fed-learning/non-dp/tuned-llama-guard3-round_4/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..553bb9d9145f7e77455bb0b76f1f25e45bf228da
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_4/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f55528c840e8f0037a342825e95c41b3c7935402adf91f7372f8e7c9b1885dc5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_4/chat_template.jinja b/fed-learning/non-dp/tuned-llama-guard3-round_4/chat_template.jinja
new file mode 100644
index 0000000000000000000000000000000000000000..8e734bc48af8bcfeecc2d0d1736179ed81008826
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_4/chat_template.jinja
@@ -0,0 +1,30 @@
+{% if messages|length % 2 == 0 %}{% set role = 'Agent' %}{% else %}{% set role = 'User' %}{% endif %}{{ " <|begin_of_text|><|start_header_id|>user<|end_header_id|>
+
+Task: Check if there is unsafe content in '" + role + "' messages in conversations according our safety policy with the below categories.
+
+<BEGIN UNSAFE CONTENT CATEGORIES>
+S1: Violent Crimes.
+S2: Non-Violent Crimes.
+S3: Sex Crimes.
+S4: Child Exploitation.
+S5: Defamation.
+S6: Specialized Advice.
+S7: Privacy.
+S8: Intellectual Property.
+S9: Indiscriminate Weapons.
+S10: Hate.
+S11: Self-Harm.
+S12: Sexual Content.
+S13: Elections.
+S14: Code Interpreter Abuse.
+<END UNSAFE CONTENT CATEGORIES>
+
+<BEGIN CONVERSATION>
+
+" }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{% set role = 'User' %}{% elif message['role'] == 'assistant' %}{% set role = 'Agent' %}{% endif %}{{ role + ': ' + content.strip() + '
+
+' }}{% endfor %}{{ "<END CONVERSATION>
+
+Provide your safety assessment for ONLY THE LAST " + role + " message in the above conversation:
+ - First line must read 'safe' or 'unsafe'.
+ - If unsafe, a second line must include a comma-separated list of violated categories.<|eot_id|><|start_header_id|>assistant<|end_header_id|>" }}
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_4/special_tokens_map.json b/fed-learning/non-dp/tuned-llama-guard3-round_4/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..92d7fab6729d723f45689803b99006fec945cc47
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_4/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b1835caa5b4d70acaa210fa222b0036f1882f9525c4660fd4810fb3e1e40ff8
+size 325
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_4/tokenizer.json b/fed-learning/non-dp/tuned-llama-guard3-round_4/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af2d22fff23798bea98b1730ae7cdacaee0b087a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_4/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ade1dac458f86f9bea8bf35b713f14e1bbed24228429534038e9f7e54ea3e8b6
+size 17208712
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_4/tokenizer_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_4/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..32408241ddf390e43ec24f8cac8608a7c66f9d1f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_4/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc2a3cb7e7ae6678f28b4a1faee31a00e0f75c475a0fd6af06a100c7f0cde0e5
+size 50550
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_5/README.md b/fed-learning/non-dp/tuned-llama-guard3-round_5/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e544b712adee647f829c2ca8c4571ab05915547a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_5/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Llama-Guard-3-8B
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_5/adapter_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_5/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..329293a4ef5d7c22bff7ad8a59dc4b7622ffc430
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_5/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a1ab765d5bfa0b7d3c35ff748fb8f1813a8150d13a764a19cb6acd2f61d77de
+size 810
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_5/adapter_model.safetensors b/fed-learning/non-dp/tuned-llama-guard3-round_5/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..553bb9d9145f7e77455bb0b76f1f25e45bf228da
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_5/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f55528c840e8f0037a342825e95c41b3c7935402adf91f7372f8e7c9b1885dc5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_5/chat_template.jinja b/fed-learning/non-dp/tuned-llama-guard3-round_5/chat_template.jinja
new file mode 100644
index 0000000000000000000000000000000000000000..8e734bc48af8bcfeecc2d0d1736179ed81008826
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_5/chat_template.jinja
@@ -0,0 +1,30 @@
+{% if messages|length % 2 == 0 %}{% set role = 'Agent' %}{% else %}{% set role = 'User' %}{% endif %}{{ " <|begin_of_text|><|start_header_id|>user<|end_header_id|>
+
+Task: Check if there is unsafe content in '" + role + "' messages in conversations according our safety policy with the below categories.
+
+<BEGIN UNSAFE CONTENT CATEGORIES>
+S1: Violent Crimes.
+S2: Non-Violent Crimes.
+S3: Sex Crimes.
+S4: Child Exploitation.
+S5: Defamation.
+S6: Specialized Advice.
+S7: Privacy.
+S8: Intellectual Property.
+S9: Indiscriminate Weapons.
+S10: Hate.
+S11: Self-Harm.
+S12: Sexual Content.
+S13: Elections.
+S14: Code Interpreter Abuse.
+<END UNSAFE CONTENT CATEGORIES>
+
+<BEGIN CONVERSATION>
+
+" }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{% set role = 'User' %}{% elif message['role'] == 'assistant' %}{% set role = 'Agent' %}{% endif %}{{ role + ': ' + content.strip() + '
+
+' }}{% endfor %}{{ "<END CONVERSATION>
+
+Provide your safety assessment for ONLY THE LAST " + role + " message in the above conversation:
+ - First line must read 'safe' or 'unsafe'.
+ - If unsafe, a second line must include a comma-separated list of violated categories.<|eot_id|><|start_header_id|>assistant<|end_header_id|>" }}
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_5/special_tokens_map.json b/fed-learning/non-dp/tuned-llama-guard3-round_5/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..92d7fab6729d723f45689803b99006fec945cc47
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_5/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b1835caa5b4d70acaa210fa222b0036f1882f9525c4660fd4810fb3e1e40ff8
+size 325
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_5/tokenizer.json b/fed-learning/non-dp/tuned-llama-guard3-round_5/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af2d22fff23798bea98b1730ae7cdacaee0b087a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_5/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ade1dac458f86f9bea8bf35b713f14e1bbed24228429534038e9f7e54ea3e8b6
+size 17208712
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_5/tokenizer_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_5/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..32408241ddf390e43ec24f8cac8608a7c66f9d1f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_5/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc2a3cb7e7ae6678f28b4a1faee31a00e0f75c475a0fd6af06a100c7f0cde0e5
+size 50550
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_6/README.md b/fed-learning/non-dp/tuned-llama-guard3-round_6/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e544b712adee647f829c2ca8c4571ab05915547a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_6/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Llama-Guard-3-8B
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_6/adapter_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_6/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..329293a4ef5d7c22bff7ad8a59dc4b7622ffc430
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_6/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a1ab765d5bfa0b7d3c35ff748fb8f1813a8150d13a764a19cb6acd2f61d77de
+size 810
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_6/adapter_model.safetensors b/fed-learning/non-dp/tuned-llama-guard3-round_6/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..553bb9d9145f7e77455bb0b76f1f25e45bf228da
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_6/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f55528c840e8f0037a342825e95c41b3c7935402adf91f7372f8e7c9b1885dc5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_6/chat_template.jinja b/fed-learning/non-dp/tuned-llama-guard3-round_6/chat_template.jinja
new file mode 100644
index 0000000000000000000000000000000000000000..8e734bc48af8bcfeecc2d0d1736179ed81008826
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_6/chat_template.jinja
@@ -0,0 +1,30 @@
+{% if messages|length % 2 == 0 %}{% set role = 'Agent' %}{% else %}{% set role = 'User' %}{% endif %}{{ " <|begin_of_text|><|start_header_id|>user<|end_header_id|>
+
+Task: Check if there is unsafe content in '" + role + "' messages in conversations according our safety policy with the below categories.
+
+<BEGIN UNSAFE CONTENT CATEGORIES>
+S1: Violent Crimes.
+S2: Non-Violent Crimes.
+S3: Sex Crimes.
+S4: Child Exploitation.
+S5: Defamation.
+S6: Specialized Advice.
+S7: Privacy.
+S8: Intellectual Property.
+S9: Indiscriminate Weapons.
+S10: Hate.
+S11: Self-Harm.
+S12: Sexual Content.
+S13: Elections.
+S14: Code Interpreter Abuse.
+<END UNSAFE CONTENT CATEGORIES>
+
+<BEGIN CONVERSATION>
+
+" }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{% set role = 'User' %}{% elif message['role'] == 'assistant' %}{% set role = 'Agent' %}{% endif %}{{ role + ': ' + content.strip() + '
+
+' }}{% endfor %}{{ "<END CONVERSATION>
+
+Provide your safety assessment for ONLY THE LAST " + role + " message in the above conversation:
+ - First line must read 'safe' or 'unsafe'.
+ - If unsafe, a second line must include a comma-separated list of violated categories.<|eot_id|><|start_header_id|>assistant<|end_header_id|>" }}
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_6/special_tokens_map.json b/fed-learning/non-dp/tuned-llama-guard3-round_6/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..92d7fab6729d723f45689803b99006fec945cc47
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_6/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b1835caa5b4d70acaa210fa222b0036f1882f9525c4660fd4810fb3e1e40ff8
+size 325
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_6/tokenizer.json b/fed-learning/non-dp/tuned-llama-guard3-round_6/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af2d22fff23798bea98b1730ae7cdacaee0b087a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_6/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ade1dac458f86f9bea8bf35b713f14e1bbed24228429534038e9f7e54ea3e8b6
+size 17208712
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_6/tokenizer_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_6/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..32408241ddf390e43ec24f8cac8608a7c66f9d1f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_6/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc2a3cb7e7ae6678f28b4a1faee31a00e0f75c475a0fd6af06a100c7f0cde0e5
+size 50550
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_7/README.md b/fed-learning/non-dp/tuned-llama-guard3-round_7/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e544b712adee647f829c2ca8c4571ab05915547a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_7/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Llama-Guard-3-8B
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_7/adapter_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_7/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..329293a4ef5d7c22bff7ad8a59dc4b7622ffc430
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_7/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a1ab765d5bfa0b7d3c35ff748fb8f1813a8150d13a764a19cb6acd2f61d77de
+size 810
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_7/adapter_model.safetensors b/fed-learning/non-dp/tuned-llama-guard3-round_7/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..553bb9d9145f7e77455bb0b76f1f25e45bf228da
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_7/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f55528c840e8f0037a342825e95c41b3c7935402adf91f7372f8e7c9b1885dc5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_7/chat_template.jinja b/fed-learning/non-dp/tuned-llama-guard3-round_7/chat_template.jinja
new file mode 100644
index 0000000000000000000000000000000000000000..8e734bc48af8bcfeecc2d0d1736179ed81008826
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_7/chat_template.jinja
@@ -0,0 +1,30 @@
+{% if messages|length % 2 == 0 %}{% set role = 'Agent' %}{% else %}{% set role = 'User' %}{% endif %}{{ " <|begin_of_text|><|start_header_id|>user<|end_header_id|>
+
+Task: Check if there is unsafe content in '" + role + "' messages in conversations according our safety policy with the below categories.
+
+<BEGIN UNSAFE CONTENT CATEGORIES>
+S1: Violent Crimes.
+S2: Non-Violent Crimes.
+S3: Sex Crimes.
+S4: Child Exploitation.
+S5: Defamation.
+S6: Specialized Advice.
+S7: Privacy.
+S8: Intellectual Property.
+S9: Indiscriminate Weapons.
+S10: Hate.
+S11: Self-Harm.
+S12: Sexual Content.
+S13: Elections.
+S14: Code Interpreter Abuse.
+<END UNSAFE CONTENT CATEGORIES>
+
+<BEGIN CONVERSATION>
+
+" }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{% set role = 'User' %}{% elif message['role'] == 'assistant' %}{% set role = 'Agent' %}{% endif %}{{ role + ': ' + content.strip() + '
+
+' }}{% endfor %}{{ "<END CONVERSATION>
+
+Provide your safety assessment for ONLY THE LAST " + role + " message in the above conversation:
+ - First line must read 'safe' or 'unsafe'.
+ - If unsafe, a second line must include a comma-separated list of violated categories.<|eot_id|><|start_header_id|>assistant<|end_header_id|>" }}
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_7/special_tokens_map.json b/fed-learning/non-dp/tuned-llama-guard3-round_7/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..92d7fab6729d723f45689803b99006fec945cc47
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_7/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b1835caa5b4d70acaa210fa222b0036f1882f9525c4660fd4810fb3e1e40ff8
+size 325
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_7/tokenizer.json b/fed-learning/non-dp/tuned-llama-guard3-round_7/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af2d22fff23798bea98b1730ae7cdacaee0b087a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_7/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ade1dac458f86f9bea8bf35b713f14e1bbed24228429534038e9f7e54ea3e8b6
+size 17208712
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_7/tokenizer_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_7/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..32408241ddf390e43ec24f8cac8608a7c66f9d1f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_7/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc2a3cb7e7ae6678f28b4a1faee31a00e0f75c475a0fd6af06a100c7f0cde0e5
+size 50550
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_8/README.md b/fed-learning/non-dp/tuned-llama-guard3-round_8/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e544b712adee647f829c2ca8c4571ab05915547a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_8/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Llama-Guard-3-8B
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_8/adapter_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_8/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..329293a4ef5d7c22bff7ad8a59dc4b7622ffc430
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_8/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a1ab765d5bfa0b7d3c35ff748fb8f1813a8150d13a764a19cb6acd2f61d77de
+size 810
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_8/adapter_model.safetensors b/fed-learning/non-dp/tuned-llama-guard3-round_8/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..553bb9d9145f7e77455bb0b76f1f25e45bf228da
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_8/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f55528c840e8f0037a342825e95c41b3c7935402adf91f7372f8e7c9b1885dc5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_8/chat_template.jinja b/fed-learning/non-dp/tuned-llama-guard3-round_8/chat_template.jinja
new file mode 100644
index 0000000000000000000000000000000000000000..8e734bc48af8bcfeecc2d0d1736179ed81008826
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_8/chat_template.jinja
@@ -0,0 +1,30 @@
+{% if messages|length % 2 == 0 %}{% set role = 'Agent' %}{% else %}{% set role = 'User' %}{% endif %}{{ " <|begin_of_text|><|start_header_id|>user<|end_header_id|>
+
+Task: Check if there is unsafe content in '" + role + "' messages in conversations according our safety policy with the below categories.
+
+<BEGIN UNSAFE CONTENT CATEGORIES>
+S1: Violent Crimes.
+S2: Non-Violent Crimes.
+S3: Sex Crimes.
+S4: Child Exploitation.
+S5: Defamation.
+S6: Specialized Advice.
+S7: Privacy.
+S8: Intellectual Property.
+S9: Indiscriminate Weapons.
+S10: Hate.
+S11: Self-Harm.
+S12: Sexual Content.
+S13: Elections.
+S14: Code Interpreter Abuse.
+<END UNSAFE CONTENT CATEGORIES>
+
+<BEGIN CONVERSATION>
+
+" }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{% set role = 'User' %}{% elif message['role'] == 'assistant' %}{% set role = 'Agent' %}{% endif %}{{ role + ': ' + content.strip() + '
+
+' }}{% endfor %}{{ "<END CONVERSATION>
+
+Provide your safety assessment for ONLY THE LAST " + role + " message in the above conversation:
+ - First line must read 'safe' or 'unsafe'.
+ - If unsafe, a second line must include a comma-separated list of violated categories.<|eot_id|><|start_header_id|>assistant<|end_header_id|>" }}
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_8/special_tokens_map.json b/fed-learning/non-dp/tuned-llama-guard3-round_8/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..92d7fab6729d723f45689803b99006fec945cc47
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_8/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b1835caa5b4d70acaa210fa222b0036f1882f9525c4660fd4810fb3e1e40ff8
+size 325
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_8/tokenizer.json b/fed-learning/non-dp/tuned-llama-guard3-round_8/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af2d22fff23798bea98b1730ae7cdacaee0b087a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_8/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ade1dac458f86f9bea8bf35b713f14e1bbed24228429534038e9f7e54ea3e8b6
+size 17208712
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_8/tokenizer_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_8/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..32408241ddf390e43ec24f8cac8608a7c66f9d1f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_8/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc2a3cb7e7ae6678f28b4a1faee31a00e0f75c475a0fd6af06a100c7f0cde0e5
+size 50550
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_9/README.md b/fed-learning/non-dp/tuned-llama-guard3-round_9/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e544b712adee647f829c2ca8c4571ab05915547a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_9/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Llama-Guard-3-8B
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_9/adapter_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_9/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..329293a4ef5d7c22bff7ad8a59dc4b7622ffc430
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_9/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a1ab765d5bfa0b7d3c35ff748fb8f1813a8150d13a764a19cb6acd2f61d77de
+size 810
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_9/adapter_model.safetensors b/fed-learning/non-dp/tuned-llama-guard3-round_9/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..553bb9d9145f7e77455bb0b76f1f25e45bf228da
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_9/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f55528c840e8f0037a342825e95c41b3c7935402adf91f7372f8e7c9b1885dc5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_9/chat_template.jinja b/fed-learning/non-dp/tuned-llama-guard3-round_9/chat_template.jinja
new file mode 100644
index 0000000000000000000000000000000000000000..8e734bc48af8bcfeecc2d0d1736179ed81008826
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_9/chat_template.jinja
@@ -0,0 +1,30 @@
+{% if messages|length % 2 == 0 %}{% set role = 'Agent' %}{% else %}{% set role = 'User' %}{% endif %}{{ " <|begin_of_text|><|start_header_id|>user<|end_header_id|>
+
+Task: Check if there is unsafe content in '" + role + "' messages in conversations according our safety policy with the below categories.
+
+<BEGIN UNSAFE CONTENT CATEGORIES>
+S1: Violent Crimes.
+S2: Non-Violent Crimes.
+S3: Sex Crimes.
+S4: Child Exploitation.
+S5: Defamation.
+S6: Specialized Advice.
+S7: Privacy.
+S8: Intellectual Property.
+S9: Indiscriminate Weapons.
+S10: Hate.
+S11: Self-Harm.
+S12: Sexual Content.
+S13: Elections.
+S14: Code Interpreter Abuse.
+<END UNSAFE CONTENT CATEGORIES>
+
+<BEGIN CONVERSATION>
+
+" }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{% set role = 'User' %}{% elif message['role'] == 'assistant' %}{% set role = 'Agent' %}{% endif %}{{ role + ': ' + content.strip() + '
+
+' }}{% endfor %}{{ "<END CONVERSATION>
+
+Provide your safety assessment for ONLY THE LAST " + role + " message in the above conversation:
+ - First line must read 'safe' or 'unsafe'.
+ - If unsafe, a second line must include a comma-separated list of violated categories.<|eot_id|><|start_header_id|>assistant<|end_header_id|>" }}
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_9/special_tokens_map.json b/fed-learning/non-dp/tuned-llama-guard3-round_9/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..92d7fab6729d723f45689803b99006fec945cc47
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_9/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b1835caa5b4d70acaa210fa222b0036f1882f9525c4660fd4810fb3e1e40ff8
+size 325
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_9/tokenizer.json b/fed-learning/non-dp/tuned-llama-guard3-round_9/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af2d22fff23798bea98b1730ae7cdacaee0b087a
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_9/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ade1dac458f86f9bea8bf35b713f14e1bbed24228429534038e9f7e54ea3e8b6
+size 17208712
diff --git a/fed-learning/non-dp/tuned-llama-guard3-round_9/tokenizer_config.json b/fed-learning/non-dp/tuned-llama-guard3-round_9/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..32408241ddf390e43ec24f8cac8608a7c66f9d1f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-llama-guard3-round_9/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc2a3cb7e7ae6678f28b4a1faee31a00e0f75c475a0fd6af06a100c7f0cde0e5
+size 50550
diff --git a/fed-learning/non-dp/tuned-md-judge-round_1/README.md b/fed-learning/non-dp/tuned-md-judge-round_1/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e13cffd9d2c7a10815128d520003acc68208b56f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_1/README.md
@@ -0,0 +1,202 @@
+---
+base_model: OpenSafetyLab/MD-Judge-v0.1
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-md-judge-round_1/adapter_config.json b/fed-learning/non-dp/tuned-md-judge-round_1/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..546e45178b8f16b3cdede31c83fae787b343c98b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_1/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6746a4999fdb0659f3c602f569a440da8f0cb25b35c6154630488f44f4359a7
+size 810
diff --git a/fed-learning/non-dp/tuned-md-judge-round_1/adapter_model.safetensors b/fed-learning/non-dp/tuned-md-judge-round_1/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..dca7d28989762c8c21d720a4d95de859ebade24f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_1/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2d254877946bdb591e3b57b04e8274697f55457b4a502838ac6961355caac99c
+size 27297032
diff --git a/fed-learning/non-dp/tuned-md-judge-round_1/special_tokens_map.json b/fed-learning/non-dp/tuned-md-judge-round_1/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..1f852aa3787fa33e54d000e81d8d7004fbfe7282
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_1/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4452b845ab9ca2afa863f59f1411a959e8a32bb4334e264b0e71d5d17dd4dcb9
+size 437
diff --git a/fed-learning/non-dp/tuned-md-judge-round_1/tokenizer.json b/fed-learning/non-dp/tuned-md-judge-round_1/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..300e828335a38f4ac67a8fd713ed8c05d5c704cf
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_1/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2699839c243202a04a90537576fc719283f638e90fe80feb469888275289575
+size 3505751
diff --git a/fed-learning/non-dp/tuned-md-judge-round_1/tokenizer.model b/fed-learning/non-dp/tuned-md-judge-round_1/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..8b443ef19c2a19acc3ac64fb9c3db4a72921dff6
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_1/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443
diff --git a/fed-learning/non-dp/tuned-md-judge-round_1/tokenizer_config.json b/fed-learning/non-dp/tuned-md-judge-round_1/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..bdfbeea6201781073c050947398f045f41c0531b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_1/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5052741b394891aa03f38b61fb141bd7ef264fefc70c189ec04b7a0a2964368
+size 1025
diff --git a/fed-learning/non-dp/tuned-md-judge-round_10/README.md b/fed-learning/non-dp/tuned-md-judge-round_10/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e13cffd9d2c7a10815128d520003acc68208b56f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_10/README.md
@@ -0,0 +1,202 @@
+---
+base_model: OpenSafetyLab/MD-Judge-v0.1
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-md-judge-round_10/adapter_config.json b/fed-learning/non-dp/tuned-md-judge-round_10/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..546e45178b8f16b3cdede31c83fae787b343c98b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_10/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6746a4999fdb0659f3c602f569a440da8f0cb25b35c6154630488f44f4359a7
+size 810
diff --git a/fed-learning/non-dp/tuned-md-judge-round_10/adapter_model.safetensors b/fed-learning/non-dp/tuned-md-judge-round_10/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..898705586029abde75d26c29a726edd1b663690f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_10/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebe1078b16cfa9ebc8d6b3b50849c428654067e7d0b5bc8f5e542246035a8be5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-md-judge-round_10/special_tokens_map.json b/fed-learning/non-dp/tuned-md-judge-round_10/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..1f852aa3787fa33e54d000e81d8d7004fbfe7282
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_10/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4452b845ab9ca2afa863f59f1411a959e8a32bb4334e264b0e71d5d17dd4dcb9
+size 437
diff --git a/fed-learning/non-dp/tuned-md-judge-round_10/tokenizer.json b/fed-learning/non-dp/tuned-md-judge-round_10/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..300e828335a38f4ac67a8fd713ed8c05d5c704cf
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_10/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2699839c243202a04a90537576fc719283f638e90fe80feb469888275289575
+size 3505751
diff --git a/fed-learning/non-dp/tuned-md-judge-round_10/tokenizer.model b/fed-learning/non-dp/tuned-md-judge-round_10/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..8b443ef19c2a19acc3ac64fb9c3db4a72921dff6
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_10/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443
diff --git a/fed-learning/non-dp/tuned-md-judge-round_10/tokenizer_config.json b/fed-learning/non-dp/tuned-md-judge-round_10/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..bdfbeea6201781073c050947398f045f41c0531b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_10/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5052741b394891aa03f38b61fb141bd7ef264fefc70c189ec04b7a0a2964368
+size 1025
diff --git a/fed-learning/non-dp/tuned-md-judge-round_11/README.md b/fed-learning/non-dp/tuned-md-judge-round_11/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e13cffd9d2c7a10815128d520003acc68208b56f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_11/README.md
@@ -0,0 +1,202 @@
+---
+base_model: OpenSafetyLab/MD-Judge-v0.1
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-md-judge-round_11/adapter_config.json b/fed-learning/non-dp/tuned-md-judge-round_11/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..546e45178b8f16b3cdede31c83fae787b343c98b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_11/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6746a4999fdb0659f3c602f569a440da8f0cb25b35c6154630488f44f4359a7
+size 810
diff --git a/fed-learning/non-dp/tuned-md-judge-round_11/adapter_model.safetensors b/fed-learning/non-dp/tuned-md-judge-round_11/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..898705586029abde75d26c29a726edd1b663690f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_11/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebe1078b16cfa9ebc8d6b3b50849c428654067e7d0b5bc8f5e542246035a8be5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-md-judge-round_11/special_tokens_map.json b/fed-learning/non-dp/tuned-md-judge-round_11/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..1f852aa3787fa33e54d000e81d8d7004fbfe7282
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_11/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4452b845ab9ca2afa863f59f1411a959e8a32bb4334e264b0e71d5d17dd4dcb9
+size 437
diff --git a/fed-learning/non-dp/tuned-md-judge-round_11/tokenizer.json b/fed-learning/non-dp/tuned-md-judge-round_11/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..300e828335a38f4ac67a8fd713ed8c05d5c704cf
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_11/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2699839c243202a04a90537576fc719283f638e90fe80feb469888275289575
+size 3505751
diff --git a/fed-learning/non-dp/tuned-md-judge-round_11/tokenizer.model b/fed-learning/non-dp/tuned-md-judge-round_11/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..8b443ef19c2a19acc3ac64fb9c3db4a72921dff6
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_11/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443
diff --git a/fed-learning/non-dp/tuned-md-judge-round_11/tokenizer_config.json b/fed-learning/non-dp/tuned-md-judge-round_11/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..bdfbeea6201781073c050947398f045f41c0531b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_11/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5052741b394891aa03f38b61fb141bd7ef264fefc70c189ec04b7a0a2964368
+size 1025
diff --git a/fed-learning/non-dp/tuned-md-judge-round_12/README.md b/fed-learning/non-dp/tuned-md-judge-round_12/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e13cffd9d2c7a10815128d520003acc68208b56f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_12/README.md
@@ -0,0 +1,202 @@
+---
+base_model: OpenSafetyLab/MD-Judge-v0.1
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-md-judge-round_12/adapter_config.json b/fed-learning/non-dp/tuned-md-judge-round_12/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..546e45178b8f16b3cdede31c83fae787b343c98b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_12/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6746a4999fdb0659f3c602f569a440da8f0cb25b35c6154630488f44f4359a7
+size 810
diff --git a/fed-learning/non-dp/tuned-md-judge-round_12/adapter_model.safetensors b/fed-learning/non-dp/tuned-md-judge-round_12/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..898705586029abde75d26c29a726edd1b663690f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_12/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebe1078b16cfa9ebc8d6b3b50849c428654067e7d0b5bc8f5e542246035a8be5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-md-judge-round_12/special_tokens_map.json b/fed-learning/non-dp/tuned-md-judge-round_12/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..1f852aa3787fa33e54d000e81d8d7004fbfe7282
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_12/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4452b845ab9ca2afa863f59f1411a959e8a32bb4334e264b0e71d5d17dd4dcb9
+size 437
diff --git a/fed-learning/non-dp/tuned-md-judge-round_12/tokenizer.json b/fed-learning/non-dp/tuned-md-judge-round_12/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..300e828335a38f4ac67a8fd713ed8c05d5c704cf
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_12/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2699839c243202a04a90537576fc719283f638e90fe80feb469888275289575
+size 3505751
diff --git a/fed-learning/non-dp/tuned-md-judge-round_12/tokenizer.model b/fed-learning/non-dp/tuned-md-judge-round_12/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..8b443ef19c2a19acc3ac64fb9c3db4a72921dff6
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_12/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443
diff --git a/fed-learning/non-dp/tuned-md-judge-round_12/tokenizer_config.json b/fed-learning/non-dp/tuned-md-judge-round_12/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..bdfbeea6201781073c050947398f045f41c0531b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_12/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5052741b394891aa03f38b61fb141bd7ef264fefc70c189ec04b7a0a2964368
+size 1025
diff --git a/fed-learning/non-dp/tuned-md-judge-round_13/README.md b/fed-learning/non-dp/tuned-md-judge-round_13/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e13cffd9d2c7a10815128d520003acc68208b56f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_13/README.md
@@ -0,0 +1,202 @@
+---
+base_model: OpenSafetyLab/MD-Judge-v0.1
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-md-judge-round_13/adapter_config.json b/fed-learning/non-dp/tuned-md-judge-round_13/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..546e45178b8f16b3cdede31c83fae787b343c98b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_13/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6746a4999fdb0659f3c602f569a440da8f0cb25b35c6154630488f44f4359a7
+size 810
diff --git a/fed-learning/non-dp/tuned-md-judge-round_13/adapter_model.safetensors b/fed-learning/non-dp/tuned-md-judge-round_13/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..898705586029abde75d26c29a726edd1b663690f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_13/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebe1078b16cfa9ebc8d6b3b50849c428654067e7d0b5bc8f5e542246035a8be5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-md-judge-round_13/special_tokens_map.json b/fed-learning/non-dp/tuned-md-judge-round_13/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..1f852aa3787fa33e54d000e81d8d7004fbfe7282
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_13/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4452b845ab9ca2afa863f59f1411a959e8a32bb4334e264b0e71d5d17dd4dcb9
+size 437
diff --git a/fed-learning/non-dp/tuned-md-judge-round_13/tokenizer.json b/fed-learning/non-dp/tuned-md-judge-round_13/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..300e828335a38f4ac67a8fd713ed8c05d5c704cf
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_13/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2699839c243202a04a90537576fc719283f638e90fe80feb469888275289575
+size 3505751
diff --git a/fed-learning/non-dp/tuned-md-judge-round_13/tokenizer.model b/fed-learning/non-dp/tuned-md-judge-round_13/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..8b443ef19c2a19acc3ac64fb9c3db4a72921dff6
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_13/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443
diff --git a/fed-learning/non-dp/tuned-md-judge-round_13/tokenizer_config.json b/fed-learning/non-dp/tuned-md-judge-round_13/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..bdfbeea6201781073c050947398f045f41c0531b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_13/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5052741b394891aa03f38b61fb141bd7ef264fefc70c189ec04b7a0a2964368
+size 1025
diff --git a/fed-learning/non-dp/tuned-md-judge-round_14/README.md b/fed-learning/non-dp/tuned-md-judge-round_14/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e13cffd9d2c7a10815128d520003acc68208b56f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_14/README.md
@@ -0,0 +1,202 @@
+---
+base_model: OpenSafetyLab/MD-Judge-v0.1
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-md-judge-round_14/adapter_config.json b/fed-learning/non-dp/tuned-md-judge-round_14/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..546e45178b8f16b3cdede31c83fae787b343c98b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_14/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6746a4999fdb0659f3c602f569a440da8f0cb25b35c6154630488f44f4359a7
+size 810
diff --git a/fed-learning/non-dp/tuned-md-judge-round_14/adapter_model.safetensors b/fed-learning/non-dp/tuned-md-judge-round_14/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..898705586029abde75d26c29a726edd1b663690f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_14/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebe1078b16cfa9ebc8d6b3b50849c428654067e7d0b5bc8f5e542246035a8be5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-md-judge-round_14/special_tokens_map.json b/fed-learning/non-dp/tuned-md-judge-round_14/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..1f852aa3787fa33e54d000e81d8d7004fbfe7282
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_14/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4452b845ab9ca2afa863f59f1411a959e8a32bb4334e264b0e71d5d17dd4dcb9
+size 437
diff --git a/fed-learning/non-dp/tuned-md-judge-round_14/tokenizer.json b/fed-learning/non-dp/tuned-md-judge-round_14/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..300e828335a38f4ac67a8fd713ed8c05d5c704cf
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_14/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2699839c243202a04a90537576fc719283f638e90fe80feb469888275289575
+size 3505751
diff --git a/fed-learning/non-dp/tuned-md-judge-round_14/tokenizer.model b/fed-learning/non-dp/tuned-md-judge-round_14/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..8b443ef19c2a19acc3ac64fb9c3db4a72921dff6
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_14/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443
diff --git a/fed-learning/non-dp/tuned-md-judge-round_14/tokenizer_config.json b/fed-learning/non-dp/tuned-md-judge-round_14/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..bdfbeea6201781073c050947398f045f41c0531b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_14/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5052741b394891aa03f38b61fb141bd7ef264fefc70c189ec04b7a0a2964368
+size 1025
diff --git a/fed-learning/non-dp/tuned-md-judge-round_15/README.md b/fed-learning/non-dp/tuned-md-judge-round_15/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e13cffd9d2c7a10815128d520003acc68208b56f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_15/README.md
@@ -0,0 +1,202 @@
+---
+base_model: OpenSafetyLab/MD-Judge-v0.1
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-md-judge-round_15/adapter_config.json b/fed-learning/non-dp/tuned-md-judge-round_15/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..546e45178b8f16b3cdede31c83fae787b343c98b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_15/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6746a4999fdb0659f3c602f569a440da8f0cb25b35c6154630488f44f4359a7
+size 810
diff --git a/fed-learning/non-dp/tuned-md-judge-round_15/adapter_model.safetensors b/fed-learning/non-dp/tuned-md-judge-round_15/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..898705586029abde75d26c29a726edd1b663690f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_15/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebe1078b16cfa9ebc8d6b3b50849c428654067e7d0b5bc8f5e542246035a8be5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-md-judge-round_15/special_tokens_map.json b/fed-learning/non-dp/tuned-md-judge-round_15/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..1f852aa3787fa33e54d000e81d8d7004fbfe7282
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_15/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4452b845ab9ca2afa863f59f1411a959e8a32bb4334e264b0e71d5d17dd4dcb9
+size 437
diff --git a/fed-learning/non-dp/tuned-md-judge-round_15/tokenizer.json b/fed-learning/non-dp/tuned-md-judge-round_15/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..300e828335a38f4ac67a8fd713ed8c05d5c704cf
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_15/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2699839c243202a04a90537576fc719283f638e90fe80feb469888275289575
+size 3505751
diff --git a/fed-learning/non-dp/tuned-md-judge-round_15/tokenizer.model b/fed-learning/non-dp/tuned-md-judge-round_15/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..8b443ef19c2a19acc3ac64fb9c3db4a72921dff6
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_15/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443
diff --git a/fed-learning/non-dp/tuned-md-judge-round_15/tokenizer_config.json b/fed-learning/non-dp/tuned-md-judge-round_15/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..bdfbeea6201781073c050947398f045f41c0531b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_15/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5052741b394891aa03f38b61fb141bd7ef264fefc70c189ec04b7a0a2964368
+size 1025
diff --git a/fed-learning/non-dp/tuned-md-judge-round_16/README.md b/fed-learning/non-dp/tuned-md-judge-round_16/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e13cffd9d2c7a10815128d520003acc68208b56f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_16/README.md
@@ -0,0 +1,202 @@
+---
+base_model: OpenSafetyLab/MD-Judge-v0.1
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-md-judge-round_16/adapter_config.json b/fed-learning/non-dp/tuned-md-judge-round_16/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..546e45178b8f16b3cdede31c83fae787b343c98b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_16/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6746a4999fdb0659f3c602f569a440da8f0cb25b35c6154630488f44f4359a7
+size 810
diff --git a/fed-learning/non-dp/tuned-md-judge-round_16/adapter_model.safetensors b/fed-learning/non-dp/tuned-md-judge-round_16/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..898705586029abde75d26c29a726edd1b663690f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_16/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebe1078b16cfa9ebc8d6b3b50849c428654067e7d0b5bc8f5e542246035a8be5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-md-judge-round_16/special_tokens_map.json b/fed-learning/non-dp/tuned-md-judge-round_16/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..1f852aa3787fa33e54d000e81d8d7004fbfe7282
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_16/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4452b845ab9ca2afa863f59f1411a959e8a32bb4334e264b0e71d5d17dd4dcb9
+size 437
diff --git a/fed-learning/non-dp/tuned-md-judge-round_16/tokenizer.json b/fed-learning/non-dp/tuned-md-judge-round_16/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..300e828335a38f4ac67a8fd713ed8c05d5c704cf
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_16/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2699839c243202a04a90537576fc719283f638e90fe80feb469888275289575
+size 3505751
diff --git a/fed-learning/non-dp/tuned-md-judge-round_16/tokenizer.model b/fed-learning/non-dp/tuned-md-judge-round_16/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..8b443ef19c2a19acc3ac64fb9c3db4a72921dff6
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_16/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443
diff --git a/fed-learning/non-dp/tuned-md-judge-round_16/tokenizer_config.json b/fed-learning/non-dp/tuned-md-judge-round_16/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..bdfbeea6201781073c050947398f045f41c0531b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_16/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5052741b394891aa03f38b61fb141bd7ef264fefc70c189ec04b7a0a2964368
+size 1025
diff --git a/fed-learning/non-dp/tuned-md-judge-round_17/README.md b/fed-learning/non-dp/tuned-md-judge-round_17/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e13cffd9d2c7a10815128d520003acc68208b56f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_17/README.md
@@ -0,0 +1,202 @@
+---
+base_model: OpenSafetyLab/MD-Judge-v0.1
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-md-judge-round_17/adapter_config.json b/fed-learning/non-dp/tuned-md-judge-round_17/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..546e45178b8f16b3cdede31c83fae787b343c98b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_17/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6746a4999fdb0659f3c602f569a440da8f0cb25b35c6154630488f44f4359a7
+size 810
diff --git a/fed-learning/non-dp/tuned-md-judge-round_17/adapter_model.safetensors b/fed-learning/non-dp/tuned-md-judge-round_17/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..898705586029abde75d26c29a726edd1b663690f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_17/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebe1078b16cfa9ebc8d6b3b50849c428654067e7d0b5bc8f5e542246035a8be5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-md-judge-round_17/special_tokens_map.json b/fed-learning/non-dp/tuned-md-judge-round_17/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..1f852aa3787fa33e54d000e81d8d7004fbfe7282
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_17/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4452b845ab9ca2afa863f59f1411a959e8a32bb4334e264b0e71d5d17dd4dcb9
+size 437
diff --git a/fed-learning/non-dp/tuned-md-judge-round_17/tokenizer.json b/fed-learning/non-dp/tuned-md-judge-round_17/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..300e828335a38f4ac67a8fd713ed8c05d5c704cf
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_17/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2699839c243202a04a90537576fc719283f638e90fe80feb469888275289575
+size 3505751
diff --git a/fed-learning/non-dp/tuned-md-judge-round_17/tokenizer.model b/fed-learning/non-dp/tuned-md-judge-round_17/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..8b443ef19c2a19acc3ac64fb9c3db4a72921dff6
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_17/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443
diff --git a/fed-learning/non-dp/tuned-md-judge-round_17/tokenizer_config.json b/fed-learning/non-dp/tuned-md-judge-round_17/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..bdfbeea6201781073c050947398f045f41c0531b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_17/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5052741b394891aa03f38b61fb141bd7ef264fefc70c189ec04b7a0a2964368
+size 1025
diff --git a/fed-learning/non-dp/tuned-md-judge-round_18/README.md b/fed-learning/non-dp/tuned-md-judge-round_18/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e13cffd9d2c7a10815128d520003acc68208b56f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_18/README.md
@@ -0,0 +1,202 @@
+---
+base_model: OpenSafetyLab/MD-Judge-v0.1
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-md-judge-round_18/adapter_config.json b/fed-learning/non-dp/tuned-md-judge-round_18/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..546e45178b8f16b3cdede31c83fae787b343c98b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_18/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6746a4999fdb0659f3c602f569a440da8f0cb25b35c6154630488f44f4359a7
+size 810
diff --git a/fed-learning/non-dp/tuned-md-judge-round_18/adapter_model.safetensors b/fed-learning/non-dp/tuned-md-judge-round_18/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..898705586029abde75d26c29a726edd1b663690f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_18/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebe1078b16cfa9ebc8d6b3b50849c428654067e7d0b5bc8f5e542246035a8be5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-md-judge-round_18/special_tokens_map.json b/fed-learning/non-dp/tuned-md-judge-round_18/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..1f852aa3787fa33e54d000e81d8d7004fbfe7282
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_18/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4452b845ab9ca2afa863f59f1411a959e8a32bb4334e264b0e71d5d17dd4dcb9
+size 437
diff --git a/fed-learning/non-dp/tuned-md-judge-round_18/tokenizer.json b/fed-learning/non-dp/tuned-md-judge-round_18/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..300e828335a38f4ac67a8fd713ed8c05d5c704cf
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_18/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2699839c243202a04a90537576fc719283f638e90fe80feb469888275289575
+size 3505751
diff --git a/fed-learning/non-dp/tuned-md-judge-round_18/tokenizer.model b/fed-learning/non-dp/tuned-md-judge-round_18/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..8b443ef19c2a19acc3ac64fb9c3db4a72921dff6
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_18/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443
diff --git a/fed-learning/non-dp/tuned-md-judge-round_18/tokenizer_config.json b/fed-learning/non-dp/tuned-md-judge-round_18/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..bdfbeea6201781073c050947398f045f41c0531b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_18/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5052741b394891aa03f38b61fb141bd7ef264fefc70c189ec04b7a0a2964368
+size 1025
diff --git a/fed-learning/non-dp/tuned-md-judge-round_19/README.md b/fed-learning/non-dp/tuned-md-judge-round_19/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e13cffd9d2c7a10815128d520003acc68208b56f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_19/README.md
@@ -0,0 +1,202 @@
+---
+base_model: OpenSafetyLab/MD-Judge-v0.1
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-md-judge-round_19/adapter_config.json b/fed-learning/non-dp/tuned-md-judge-round_19/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..546e45178b8f16b3cdede31c83fae787b343c98b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_19/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6746a4999fdb0659f3c602f569a440da8f0cb25b35c6154630488f44f4359a7
+size 810
diff --git a/fed-learning/non-dp/tuned-md-judge-round_19/adapter_model.safetensors b/fed-learning/non-dp/tuned-md-judge-round_19/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..898705586029abde75d26c29a726edd1b663690f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_19/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebe1078b16cfa9ebc8d6b3b50849c428654067e7d0b5bc8f5e542246035a8be5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-md-judge-round_19/special_tokens_map.json b/fed-learning/non-dp/tuned-md-judge-round_19/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..1f852aa3787fa33e54d000e81d8d7004fbfe7282
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_19/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4452b845ab9ca2afa863f59f1411a959e8a32bb4334e264b0e71d5d17dd4dcb9
+size 437
diff --git a/fed-learning/non-dp/tuned-md-judge-round_19/tokenizer.json b/fed-learning/non-dp/tuned-md-judge-round_19/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..300e828335a38f4ac67a8fd713ed8c05d5c704cf
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_19/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2699839c243202a04a90537576fc719283f638e90fe80feb469888275289575
+size 3505751
diff --git a/fed-learning/non-dp/tuned-md-judge-round_19/tokenizer.model b/fed-learning/non-dp/tuned-md-judge-round_19/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..8b443ef19c2a19acc3ac64fb9c3db4a72921dff6
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_19/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443
diff --git a/fed-learning/non-dp/tuned-md-judge-round_19/tokenizer_config.json b/fed-learning/non-dp/tuned-md-judge-round_19/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..bdfbeea6201781073c050947398f045f41c0531b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_19/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5052741b394891aa03f38b61fb141bd7ef264fefc70c189ec04b7a0a2964368
+size 1025
diff --git a/fed-learning/non-dp/tuned-md-judge-round_2/README.md b/fed-learning/non-dp/tuned-md-judge-round_2/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e13cffd9d2c7a10815128d520003acc68208b56f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_2/README.md
@@ -0,0 +1,202 @@
+---
+base_model: OpenSafetyLab/MD-Judge-v0.1
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-md-judge-round_2/adapter_config.json b/fed-learning/non-dp/tuned-md-judge-round_2/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..546e45178b8f16b3cdede31c83fae787b343c98b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_2/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6746a4999fdb0659f3c602f569a440da8f0cb25b35c6154630488f44f4359a7
+size 810
diff --git a/fed-learning/non-dp/tuned-md-judge-round_2/adapter_model.safetensors b/fed-learning/non-dp/tuned-md-judge-round_2/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..65479a866f05c89f99625e2382b95debcc7f038c
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_2/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:248fddcc26fe3e5242682427c2767fd94f1edd57697594888ad527dcae7d3f07
+size 27297032
diff --git a/fed-learning/non-dp/tuned-md-judge-round_2/special_tokens_map.json b/fed-learning/non-dp/tuned-md-judge-round_2/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..1f852aa3787fa33e54d000e81d8d7004fbfe7282
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_2/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4452b845ab9ca2afa863f59f1411a959e8a32bb4334e264b0e71d5d17dd4dcb9
+size 437
diff --git a/fed-learning/non-dp/tuned-md-judge-round_2/tokenizer.json b/fed-learning/non-dp/tuned-md-judge-round_2/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..300e828335a38f4ac67a8fd713ed8c05d5c704cf
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_2/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2699839c243202a04a90537576fc719283f638e90fe80feb469888275289575
+size 3505751
diff --git a/fed-learning/non-dp/tuned-md-judge-round_2/tokenizer.model b/fed-learning/non-dp/tuned-md-judge-round_2/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..8b443ef19c2a19acc3ac64fb9c3db4a72921dff6
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_2/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443
diff --git a/fed-learning/non-dp/tuned-md-judge-round_2/tokenizer_config.json b/fed-learning/non-dp/tuned-md-judge-round_2/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..bdfbeea6201781073c050947398f045f41c0531b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_2/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5052741b394891aa03f38b61fb141bd7ef264fefc70c189ec04b7a0a2964368
+size 1025
diff --git a/fed-learning/non-dp/tuned-md-judge-round_20/README.md b/fed-learning/non-dp/tuned-md-judge-round_20/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e13cffd9d2c7a10815128d520003acc68208b56f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_20/README.md
@@ -0,0 +1,202 @@
+---
+base_model: OpenSafetyLab/MD-Judge-v0.1
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-md-judge-round_20/adapter_config.json b/fed-learning/non-dp/tuned-md-judge-round_20/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..546e45178b8f16b3cdede31c83fae787b343c98b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_20/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6746a4999fdb0659f3c602f569a440da8f0cb25b35c6154630488f44f4359a7
+size 810
diff --git a/fed-learning/non-dp/tuned-md-judge-round_20/adapter_model.safetensors b/fed-learning/non-dp/tuned-md-judge-round_20/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..898705586029abde75d26c29a726edd1b663690f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_20/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebe1078b16cfa9ebc8d6b3b50849c428654067e7d0b5bc8f5e542246035a8be5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-md-judge-round_20/special_tokens_map.json b/fed-learning/non-dp/tuned-md-judge-round_20/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..1f852aa3787fa33e54d000e81d8d7004fbfe7282
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_20/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4452b845ab9ca2afa863f59f1411a959e8a32bb4334e264b0e71d5d17dd4dcb9
+size 437
diff --git a/fed-learning/non-dp/tuned-md-judge-round_20/tokenizer.json b/fed-learning/non-dp/tuned-md-judge-round_20/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..300e828335a38f4ac67a8fd713ed8c05d5c704cf
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_20/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2699839c243202a04a90537576fc719283f638e90fe80feb469888275289575
+size 3505751
diff --git a/fed-learning/non-dp/tuned-md-judge-round_20/tokenizer.model b/fed-learning/non-dp/tuned-md-judge-round_20/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..8b443ef19c2a19acc3ac64fb9c3db4a72921dff6
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_20/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443
diff --git a/fed-learning/non-dp/tuned-md-judge-round_20/tokenizer_config.json b/fed-learning/non-dp/tuned-md-judge-round_20/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..bdfbeea6201781073c050947398f045f41c0531b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_20/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5052741b394891aa03f38b61fb141bd7ef264fefc70c189ec04b7a0a2964368
+size 1025
diff --git a/fed-learning/non-dp/tuned-md-judge-round_21/README.md b/fed-learning/non-dp/tuned-md-judge-round_21/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e13cffd9d2c7a10815128d520003acc68208b56f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_21/README.md
@@ -0,0 +1,202 @@
+---
+base_model: OpenSafetyLab/MD-Judge-v0.1
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-md-judge-round_21/adapter_config.json b/fed-learning/non-dp/tuned-md-judge-round_21/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..546e45178b8f16b3cdede31c83fae787b343c98b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_21/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6746a4999fdb0659f3c602f569a440da8f0cb25b35c6154630488f44f4359a7
+size 810
diff --git a/fed-learning/non-dp/tuned-md-judge-round_21/adapter_model.safetensors b/fed-learning/non-dp/tuned-md-judge-round_21/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..898705586029abde75d26c29a726edd1b663690f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_21/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebe1078b16cfa9ebc8d6b3b50849c428654067e7d0b5bc8f5e542246035a8be5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-md-judge-round_21/special_tokens_map.json b/fed-learning/non-dp/tuned-md-judge-round_21/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..1f852aa3787fa33e54d000e81d8d7004fbfe7282
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_21/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4452b845ab9ca2afa863f59f1411a959e8a32bb4334e264b0e71d5d17dd4dcb9
+size 437
diff --git a/fed-learning/non-dp/tuned-md-judge-round_21/tokenizer.json b/fed-learning/non-dp/tuned-md-judge-round_21/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..300e828335a38f4ac67a8fd713ed8c05d5c704cf
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_21/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2699839c243202a04a90537576fc719283f638e90fe80feb469888275289575
+size 3505751
diff --git a/fed-learning/non-dp/tuned-md-judge-round_21/tokenizer.model b/fed-learning/non-dp/tuned-md-judge-round_21/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..8b443ef19c2a19acc3ac64fb9c3db4a72921dff6
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_21/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443
diff --git a/fed-learning/non-dp/tuned-md-judge-round_21/tokenizer_config.json b/fed-learning/non-dp/tuned-md-judge-round_21/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..bdfbeea6201781073c050947398f045f41c0531b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_21/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5052741b394891aa03f38b61fb141bd7ef264fefc70c189ec04b7a0a2964368
+size 1025
diff --git a/fed-learning/non-dp/tuned-md-judge-round_22/README.md b/fed-learning/non-dp/tuned-md-judge-round_22/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e13cffd9d2c7a10815128d520003acc68208b56f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_22/README.md
@@ -0,0 +1,202 @@
+---
+base_model: OpenSafetyLab/MD-Judge-v0.1
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-md-judge-round_22/adapter_config.json b/fed-learning/non-dp/tuned-md-judge-round_22/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..546e45178b8f16b3cdede31c83fae787b343c98b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_22/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6746a4999fdb0659f3c602f569a440da8f0cb25b35c6154630488f44f4359a7
+size 810
diff --git a/fed-learning/non-dp/tuned-md-judge-round_22/adapter_model.safetensors b/fed-learning/non-dp/tuned-md-judge-round_22/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..898705586029abde75d26c29a726edd1b663690f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_22/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebe1078b16cfa9ebc8d6b3b50849c428654067e7d0b5bc8f5e542246035a8be5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-md-judge-round_22/special_tokens_map.json b/fed-learning/non-dp/tuned-md-judge-round_22/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..1f852aa3787fa33e54d000e81d8d7004fbfe7282
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_22/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4452b845ab9ca2afa863f59f1411a959e8a32bb4334e264b0e71d5d17dd4dcb9
+size 437
diff --git a/fed-learning/non-dp/tuned-md-judge-round_22/tokenizer.json b/fed-learning/non-dp/tuned-md-judge-round_22/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..300e828335a38f4ac67a8fd713ed8c05d5c704cf
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_22/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2699839c243202a04a90537576fc719283f638e90fe80feb469888275289575
+size 3505751
diff --git a/fed-learning/non-dp/tuned-md-judge-round_22/tokenizer.model b/fed-learning/non-dp/tuned-md-judge-round_22/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..8b443ef19c2a19acc3ac64fb9c3db4a72921dff6
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_22/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443
diff --git a/fed-learning/non-dp/tuned-md-judge-round_22/tokenizer_config.json b/fed-learning/non-dp/tuned-md-judge-round_22/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..bdfbeea6201781073c050947398f045f41c0531b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_22/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5052741b394891aa03f38b61fb141bd7ef264fefc70c189ec04b7a0a2964368
+size 1025
diff --git a/fed-learning/non-dp/tuned-md-judge-round_23/README.md b/fed-learning/non-dp/tuned-md-judge-round_23/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e13cffd9d2c7a10815128d520003acc68208b56f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_23/README.md
@@ -0,0 +1,202 @@
+---
+base_model: OpenSafetyLab/MD-Judge-v0.1
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-md-judge-round_23/adapter_config.json b/fed-learning/non-dp/tuned-md-judge-round_23/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..546e45178b8f16b3cdede31c83fae787b343c98b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_23/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6746a4999fdb0659f3c602f569a440da8f0cb25b35c6154630488f44f4359a7
+size 810
diff --git a/fed-learning/non-dp/tuned-md-judge-round_23/adapter_model.safetensors b/fed-learning/non-dp/tuned-md-judge-round_23/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..898705586029abde75d26c29a726edd1b663690f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_23/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebe1078b16cfa9ebc8d6b3b50849c428654067e7d0b5bc8f5e542246035a8be5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-md-judge-round_23/special_tokens_map.json b/fed-learning/non-dp/tuned-md-judge-round_23/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..1f852aa3787fa33e54d000e81d8d7004fbfe7282
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_23/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4452b845ab9ca2afa863f59f1411a959e8a32bb4334e264b0e71d5d17dd4dcb9
+size 437
diff --git a/fed-learning/non-dp/tuned-md-judge-round_23/tokenizer.json b/fed-learning/non-dp/tuned-md-judge-round_23/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..300e828335a38f4ac67a8fd713ed8c05d5c704cf
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_23/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2699839c243202a04a90537576fc719283f638e90fe80feb469888275289575
+size 3505751
diff --git a/fed-learning/non-dp/tuned-md-judge-round_23/tokenizer.model b/fed-learning/non-dp/tuned-md-judge-round_23/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..8b443ef19c2a19acc3ac64fb9c3db4a72921dff6
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_23/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443
diff --git a/fed-learning/non-dp/tuned-md-judge-round_23/tokenizer_config.json b/fed-learning/non-dp/tuned-md-judge-round_23/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..bdfbeea6201781073c050947398f045f41c0531b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_23/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5052741b394891aa03f38b61fb141bd7ef264fefc70c189ec04b7a0a2964368
+size 1025
diff --git a/fed-learning/non-dp/tuned-md-judge-round_24/README.md b/fed-learning/non-dp/tuned-md-judge-round_24/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e13cffd9d2c7a10815128d520003acc68208b56f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_24/README.md
@@ -0,0 +1,202 @@
+---
+base_model: OpenSafetyLab/MD-Judge-v0.1
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-md-judge-round_24/adapter_config.json b/fed-learning/non-dp/tuned-md-judge-round_24/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..546e45178b8f16b3cdede31c83fae787b343c98b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_24/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6746a4999fdb0659f3c602f569a440da8f0cb25b35c6154630488f44f4359a7
+size 810
diff --git a/fed-learning/non-dp/tuned-md-judge-round_24/adapter_model.safetensors b/fed-learning/non-dp/tuned-md-judge-round_24/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..898705586029abde75d26c29a726edd1b663690f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_24/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebe1078b16cfa9ebc8d6b3b50849c428654067e7d0b5bc8f5e542246035a8be5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-md-judge-round_24/special_tokens_map.json b/fed-learning/non-dp/tuned-md-judge-round_24/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..1f852aa3787fa33e54d000e81d8d7004fbfe7282
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_24/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4452b845ab9ca2afa863f59f1411a959e8a32bb4334e264b0e71d5d17dd4dcb9
+size 437
diff --git a/fed-learning/non-dp/tuned-md-judge-round_24/tokenizer.json b/fed-learning/non-dp/tuned-md-judge-round_24/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..300e828335a38f4ac67a8fd713ed8c05d5c704cf
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_24/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2699839c243202a04a90537576fc719283f638e90fe80feb469888275289575
+size 3505751
diff --git a/fed-learning/non-dp/tuned-md-judge-round_24/tokenizer.model b/fed-learning/non-dp/tuned-md-judge-round_24/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..8b443ef19c2a19acc3ac64fb9c3db4a72921dff6
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_24/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443
diff --git a/fed-learning/non-dp/tuned-md-judge-round_24/tokenizer_config.json b/fed-learning/non-dp/tuned-md-judge-round_24/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..bdfbeea6201781073c050947398f045f41c0531b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_24/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5052741b394891aa03f38b61fb141bd7ef264fefc70c189ec04b7a0a2964368
+size 1025
diff --git a/fed-learning/non-dp/tuned-md-judge-round_25/README.md b/fed-learning/non-dp/tuned-md-judge-round_25/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e13cffd9d2c7a10815128d520003acc68208b56f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_25/README.md
@@ -0,0 +1,202 @@
+---
+base_model: OpenSafetyLab/MD-Judge-v0.1
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-md-judge-round_25/adapter_config.json b/fed-learning/non-dp/tuned-md-judge-round_25/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..546e45178b8f16b3cdede31c83fae787b343c98b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_25/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6746a4999fdb0659f3c602f569a440da8f0cb25b35c6154630488f44f4359a7
+size 810
diff --git a/fed-learning/non-dp/tuned-md-judge-round_25/adapter_model.safetensors b/fed-learning/non-dp/tuned-md-judge-round_25/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..898705586029abde75d26c29a726edd1b663690f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_25/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebe1078b16cfa9ebc8d6b3b50849c428654067e7d0b5bc8f5e542246035a8be5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-md-judge-round_25/special_tokens_map.json b/fed-learning/non-dp/tuned-md-judge-round_25/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..1f852aa3787fa33e54d000e81d8d7004fbfe7282
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_25/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4452b845ab9ca2afa863f59f1411a959e8a32bb4334e264b0e71d5d17dd4dcb9
+size 437
diff --git a/fed-learning/non-dp/tuned-md-judge-round_25/tokenizer.json b/fed-learning/non-dp/tuned-md-judge-round_25/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..300e828335a38f4ac67a8fd713ed8c05d5c704cf
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_25/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2699839c243202a04a90537576fc719283f638e90fe80feb469888275289575
+size 3505751
diff --git a/fed-learning/non-dp/tuned-md-judge-round_25/tokenizer.model b/fed-learning/non-dp/tuned-md-judge-round_25/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..8b443ef19c2a19acc3ac64fb9c3db4a72921dff6
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_25/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443
diff --git a/fed-learning/non-dp/tuned-md-judge-round_25/tokenizer_config.json b/fed-learning/non-dp/tuned-md-judge-round_25/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..bdfbeea6201781073c050947398f045f41c0531b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_25/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5052741b394891aa03f38b61fb141bd7ef264fefc70c189ec04b7a0a2964368
+size 1025
diff --git a/fed-learning/non-dp/tuned-md-judge-round_26/README.md b/fed-learning/non-dp/tuned-md-judge-round_26/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e13cffd9d2c7a10815128d520003acc68208b56f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_26/README.md
@@ -0,0 +1,202 @@
+---
+base_model: OpenSafetyLab/MD-Judge-v0.1
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-md-judge-round_26/adapter_config.json b/fed-learning/non-dp/tuned-md-judge-round_26/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..546e45178b8f16b3cdede31c83fae787b343c98b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_26/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6746a4999fdb0659f3c602f569a440da8f0cb25b35c6154630488f44f4359a7
+size 810
diff --git a/fed-learning/non-dp/tuned-md-judge-round_26/adapter_model.safetensors b/fed-learning/non-dp/tuned-md-judge-round_26/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..898705586029abde75d26c29a726edd1b663690f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_26/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebe1078b16cfa9ebc8d6b3b50849c428654067e7d0b5bc8f5e542246035a8be5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-md-judge-round_26/special_tokens_map.json b/fed-learning/non-dp/tuned-md-judge-round_26/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..1f852aa3787fa33e54d000e81d8d7004fbfe7282
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_26/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4452b845ab9ca2afa863f59f1411a959e8a32bb4334e264b0e71d5d17dd4dcb9
+size 437
diff --git a/fed-learning/non-dp/tuned-md-judge-round_26/tokenizer.json b/fed-learning/non-dp/tuned-md-judge-round_26/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..300e828335a38f4ac67a8fd713ed8c05d5c704cf
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_26/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2699839c243202a04a90537576fc719283f638e90fe80feb469888275289575
+size 3505751
diff --git a/fed-learning/non-dp/tuned-md-judge-round_26/tokenizer.model b/fed-learning/non-dp/tuned-md-judge-round_26/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..8b443ef19c2a19acc3ac64fb9c3db4a72921dff6
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_26/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443
diff --git a/fed-learning/non-dp/tuned-md-judge-round_26/tokenizer_config.json b/fed-learning/non-dp/tuned-md-judge-round_26/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..bdfbeea6201781073c050947398f045f41c0531b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_26/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5052741b394891aa03f38b61fb141bd7ef264fefc70c189ec04b7a0a2964368
+size 1025
diff --git a/fed-learning/non-dp/tuned-md-judge-round_27/README.md b/fed-learning/non-dp/tuned-md-judge-round_27/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e13cffd9d2c7a10815128d520003acc68208b56f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_27/README.md
@@ -0,0 +1,202 @@
+---
+base_model: OpenSafetyLab/MD-Judge-v0.1
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-md-judge-round_27/adapter_config.json b/fed-learning/non-dp/tuned-md-judge-round_27/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..546e45178b8f16b3cdede31c83fae787b343c98b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_27/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6746a4999fdb0659f3c602f569a440da8f0cb25b35c6154630488f44f4359a7
+size 810
diff --git a/fed-learning/non-dp/tuned-md-judge-round_27/adapter_model.safetensors b/fed-learning/non-dp/tuned-md-judge-round_27/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..898705586029abde75d26c29a726edd1b663690f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_27/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebe1078b16cfa9ebc8d6b3b50849c428654067e7d0b5bc8f5e542246035a8be5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-md-judge-round_27/special_tokens_map.json b/fed-learning/non-dp/tuned-md-judge-round_27/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..1f852aa3787fa33e54d000e81d8d7004fbfe7282
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_27/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4452b845ab9ca2afa863f59f1411a959e8a32bb4334e264b0e71d5d17dd4dcb9
+size 437
diff --git a/fed-learning/non-dp/tuned-md-judge-round_27/tokenizer.json b/fed-learning/non-dp/tuned-md-judge-round_27/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..300e828335a38f4ac67a8fd713ed8c05d5c704cf
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_27/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2699839c243202a04a90537576fc719283f638e90fe80feb469888275289575
+size 3505751
diff --git a/fed-learning/non-dp/tuned-md-judge-round_27/tokenizer.model b/fed-learning/non-dp/tuned-md-judge-round_27/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..8b443ef19c2a19acc3ac64fb9c3db4a72921dff6
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_27/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443
diff --git a/fed-learning/non-dp/tuned-md-judge-round_27/tokenizer_config.json b/fed-learning/non-dp/tuned-md-judge-round_27/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..bdfbeea6201781073c050947398f045f41c0531b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_27/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5052741b394891aa03f38b61fb141bd7ef264fefc70c189ec04b7a0a2964368
+size 1025
diff --git a/fed-learning/non-dp/tuned-md-judge-round_28/README.md b/fed-learning/non-dp/tuned-md-judge-round_28/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e13cffd9d2c7a10815128d520003acc68208b56f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_28/README.md
@@ -0,0 +1,202 @@
+---
+base_model: OpenSafetyLab/MD-Judge-v0.1
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-md-judge-round_28/adapter_config.json b/fed-learning/non-dp/tuned-md-judge-round_28/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..546e45178b8f16b3cdede31c83fae787b343c98b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_28/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6746a4999fdb0659f3c602f569a440da8f0cb25b35c6154630488f44f4359a7
+size 810
diff --git a/fed-learning/non-dp/tuned-md-judge-round_28/adapter_model.safetensors b/fed-learning/non-dp/tuned-md-judge-round_28/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..898705586029abde75d26c29a726edd1b663690f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_28/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebe1078b16cfa9ebc8d6b3b50849c428654067e7d0b5bc8f5e542246035a8be5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-md-judge-round_28/special_tokens_map.json b/fed-learning/non-dp/tuned-md-judge-round_28/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..1f852aa3787fa33e54d000e81d8d7004fbfe7282
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_28/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4452b845ab9ca2afa863f59f1411a959e8a32bb4334e264b0e71d5d17dd4dcb9
+size 437
diff --git a/fed-learning/non-dp/tuned-md-judge-round_28/tokenizer.json b/fed-learning/non-dp/tuned-md-judge-round_28/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..300e828335a38f4ac67a8fd713ed8c05d5c704cf
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_28/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2699839c243202a04a90537576fc719283f638e90fe80feb469888275289575
+size 3505751
diff --git a/fed-learning/non-dp/tuned-md-judge-round_28/tokenizer.model b/fed-learning/non-dp/tuned-md-judge-round_28/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..8b443ef19c2a19acc3ac64fb9c3db4a72921dff6
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_28/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443
diff --git a/fed-learning/non-dp/tuned-md-judge-round_28/tokenizer_config.json b/fed-learning/non-dp/tuned-md-judge-round_28/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..bdfbeea6201781073c050947398f045f41c0531b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_28/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5052741b394891aa03f38b61fb141bd7ef264fefc70c189ec04b7a0a2964368
+size 1025
diff --git a/fed-learning/non-dp/tuned-md-judge-round_29/README.md b/fed-learning/non-dp/tuned-md-judge-round_29/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e13cffd9d2c7a10815128d520003acc68208b56f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_29/README.md
@@ -0,0 +1,202 @@
+---
+base_model: OpenSafetyLab/MD-Judge-v0.1
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-md-judge-round_29/adapter_config.json b/fed-learning/non-dp/tuned-md-judge-round_29/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..546e45178b8f16b3cdede31c83fae787b343c98b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_29/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6746a4999fdb0659f3c602f569a440da8f0cb25b35c6154630488f44f4359a7
+size 810
diff --git a/fed-learning/non-dp/tuned-md-judge-round_29/adapter_model.safetensors b/fed-learning/non-dp/tuned-md-judge-round_29/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..898705586029abde75d26c29a726edd1b663690f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_29/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebe1078b16cfa9ebc8d6b3b50849c428654067e7d0b5bc8f5e542246035a8be5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-md-judge-round_29/special_tokens_map.json b/fed-learning/non-dp/tuned-md-judge-round_29/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..1f852aa3787fa33e54d000e81d8d7004fbfe7282
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_29/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4452b845ab9ca2afa863f59f1411a959e8a32bb4334e264b0e71d5d17dd4dcb9
+size 437
diff --git a/fed-learning/non-dp/tuned-md-judge-round_29/tokenizer.json b/fed-learning/non-dp/tuned-md-judge-round_29/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..300e828335a38f4ac67a8fd713ed8c05d5c704cf
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_29/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2699839c243202a04a90537576fc719283f638e90fe80feb469888275289575
+size 3505751
diff --git a/fed-learning/non-dp/tuned-md-judge-round_29/tokenizer.model b/fed-learning/non-dp/tuned-md-judge-round_29/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..8b443ef19c2a19acc3ac64fb9c3db4a72921dff6
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_29/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443
diff --git a/fed-learning/non-dp/tuned-md-judge-round_29/tokenizer_config.json b/fed-learning/non-dp/tuned-md-judge-round_29/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..bdfbeea6201781073c050947398f045f41c0531b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_29/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5052741b394891aa03f38b61fb141bd7ef264fefc70c189ec04b7a0a2964368
+size 1025
diff --git a/fed-learning/non-dp/tuned-md-judge-round_3/README.md b/fed-learning/non-dp/tuned-md-judge-round_3/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e13cffd9d2c7a10815128d520003acc68208b56f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_3/README.md
@@ -0,0 +1,202 @@
+---
+base_model: OpenSafetyLab/MD-Judge-v0.1
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-md-judge-round_3/adapter_config.json b/fed-learning/non-dp/tuned-md-judge-round_3/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..546e45178b8f16b3cdede31c83fae787b343c98b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_3/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6746a4999fdb0659f3c602f569a440da8f0cb25b35c6154630488f44f4359a7
+size 810
diff --git a/fed-learning/non-dp/tuned-md-judge-round_3/adapter_model.safetensors b/fed-learning/non-dp/tuned-md-judge-round_3/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..898705586029abde75d26c29a726edd1b663690f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_3/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebe1078b16cfa9ebc8d6b3b50849c428654067e7d0b5bc8f5e542246035a8be5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-md-judge-round_3/special_tokens_map.json b/fed-learning/non-dp/tuned-md-judge-round_3/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..1f852aa3787fa33e54d000e81d8d7004fbfe7282
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_3/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4452b845ab9ca2afa863f59f1411a959e8a32bb4334e264b0e71d5d17dd4dcb9
+size 437
diff --git a/fed-learning/non-dp/tuned-md-judge-round_3/tokenizer.json b/fed-learning/non-dp/tuned-md-judge-round_3/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..300e828335a38f4ac67a8fd713ed8c05d5c704cf
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_3/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2699839c243202a04a90537576fc719283f638e90fe80feb469888275289575
+size 3505751
diff --git a/fed-learning/non-dp/tuned-md-judge-round_3/tokenizer.model b/fed-learning/non-dp/tuned-md-judge-round_3/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..8b443ef19c2a19acc3ac64fb9c3db4a72921dff6
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_3/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443
diff --git a/fed-learning/non-dp/tuned-md-judge-round_3/tokenizer_config.json b/fed-learning/non-dp/tuned-md-judge-round_3/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..bdfbeea6201781073c050947398f045f41c0531b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_3/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5052741b394891aa03f38b61fb141bd7ef264fefc70c189ec04b7a0a2964368
+size 1025
diff --git a/fed-learning/non-dp/tuned-md-judge-round_30/README.md b/fed-learning/non-dp/tuned-md-judge-round_30/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e13cffd9d2c7a10815128d520003acc68208b56f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_30/README.md
@@ -0,0 +1,202 @@
+---
+base_model: OpenSafetyLab/MD-Judge-v0.1
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-md-judge-round_30/adapter_config.json b/fed-learning/non-dp/tuned-md-judge-round_30/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..546e45178b8f16b3cdede31c83fae787b343c98b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_30/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6746a4999fdb0659f3c602f569a440da8f0cb25b35c6154630488f44f4359a7
+size 810
diff --git a/fed-learning/non-dp/tuned-md-judge-round_30/adapter_model.safetensors b/fed-learning/non-dp/tuned-md-judge-round_30/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..898705586029abde75d26c29a726edd1b663690f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_30/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebe1078b16cfa9ebc8d6b3b50849c428654067e7d0b5bc8f5e542246035a8be5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-md-judge-round_30/special_tokens_map.json b/fed-learning/non-dp/tuned-md-judge-round_30/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..1f852aa3787fa33e54d000e81d8d7004fbfe7282
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_30/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4452b845ab9ca2afa863f59f1411a959e8a32bb4334e264b0e71d5d17dd4dcb9
+size 437
diff --git a/fed-learning/non-dp/tuned-md-judge-round_30/tokenizer.json b/fed-learning/non-dp/tuned-md-judge-round_30/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..300e828335a38f4ac67a8fd713ed8c05d5c704cf
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_30/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2699839c243202a04a90537576fc719283f638e90fe80feb469888275289575
+size 3505751
diff --git a/fed-learning/non-dp/tuned-md-judge-round_30/tokenizer.model b/fed-learning/non-dp/tuned-md-judge-round_30/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..8b443ef19c2a19acc3ac64fb9c3db4a72921dff6
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_30/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443
diff --git a/fed-learning/non-dp/tuned-md-judge-round_30/tokenizer_config.json b/fed-learning/non-dp/tuned-md-judge-round_30/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..bdfbeea6201781073c050947398f045f41c0531b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_30/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5052741b394891aa03f38b61fb141bd7ef264fefc70c189ec04b7a0a2964368
+size 1025
diff --git a/fed-learning/non-dp/tuned-md-judge-round_4/README.md b/fed-learning/non-dp/tuned-md-judge-round_4/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e13cffd9d2c7a10815128d520003acc68208b56f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_4/README.md
@@ -0,0 +1,202 @@
+---
+base_model: OpenSafetyLab/MD-Judge-v0.1
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-md-judge-round_4/adapter_config.json b/fed-learning/non-dp/tuned-md-judge-round_4/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..546e45178b8f16b3cdede31c83fae787b343c98b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_4/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6746a4999fdb0659f3c602f569a440da8f0cb25b35c6154630488f44f4359a7
+size 810
diff --git a/fed-learning/non-dp/tuned-md-judge-round_4/adapter_model.safetensors b/fed-learning/non-dp/tuned-md-judge-round_4/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..898705586029abde75d26c29a726edd1b663690f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_4/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebe1078b16cfa9ebc8d6b3b50849c428654067e7d0b5bc8f5e542246035a8be5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-md-judge-round_4/special_tokens_map.json b/fed-learning/non-dp/tuned-md-judge-round_4/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..1f852aa3787fa33e54d000e81d8d7004fbfe7282
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_4/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4452b845ab9ca2afa863f59f1411a959e8a32bb4334e264b0e71d5d17dd4dcb9
+size 437
diff --git a/fed-learning/non-dp/tuned-md-judge-round_4/tokenizer.json b/fed-learning/non-dp/tuned-md-judge-round_4/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..300e828335a38f4ac67a8fd713ed8c05d5c704cf
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_4/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2699839c243202a04a90537576fc719283f638e90fe80feb469888275289575
+size 3505751
diff --git a/fed-learning/non-dp/tuned-md-judge-round_4/tokenizer.model b/fed-learning/non-dp/tuned-md-judge-round_4/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..8b443ef19c2a19acc3ac64fb9c3db4a72921dff6
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_4/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443
diff --git a/fed-learning/non-dp/tuned-md-judge-round_4/tokenizer_config.json b/fed-learning/non-dp/tuned-md-judge-round_4/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..bdfbeea6201781073c050947398f045f41c0531b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_4/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5052741b394891aa03f38b61fb141bd7ef264fefc70c189ec04b7a0a2964368
+size 1025
diff --git a/fed-learning/non-dp/tuned-md-judge-round_5/README.md b/fed-learning/non-dp/tuned-md-judge-round_5/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e13cffd9d2c7a10815128d520003acc68208b56f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_5/README.md
@@ -0,0 +1,202 @@
+---
+base_model: OpenSafetyLab/MD-Judge-v0.1
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-md-judge-round_5/adapter_config.json b/fed-learning/non-dp/tuned-md-judge-round_5/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..546e45178b8f16b3cdede31c83fae787b343c98b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_5/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6746a4999fdb0659f3c602f569a440da8f0cb25b35c6154630488f44f4359a7
+size 810
diff --git a/fed-learning/non-dp/tuned-md-judge-round_5/adapter_model.safetensors b/fed-learning/non-dp/tuned-md-judge-round_5/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..898705586029abde75d26c29a726edd1b663690f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_5/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebe1078b16cfa9ebc8d6b3b50849c428654067e7d0b5bc8f5e542246035a8be5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-md-judge-round_5/special_tokens_map.json b/fed-learning/non-dp/tuned-md-judge-round_5/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..1f852aa3787fa33e54d000e81d8d7004fbfe7282
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_5/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4452b845ab9ca2afa863f59f1411a959e8a32bb4334e264b0e71d5d17dd4dcb9
+size 437
diff --git a/fed-learning/non-dp/tuned-md-judge-round_5/tokenizer.json b/fed-learning/non-dp/tuned-md-judge-round_5/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..300e828335a38f4ac67a8fd713ed8c05d5c704cf
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_5/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2699839c243202a04a90537576fc719283f638e90fe80feb469888275289575
+size 3505751
diff --git a/fed-learning/non-dp/tuned-md-judge-round_5/tokenizer.model b/fed-learning/non-dp/tuned-md-judge-round_5/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..8b443ef19c2a19acc3ac64fb9c3db4a72921dff6
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_5/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443
diff --git a/fed-learning/non-dp/tuned-md-judge-round_5/tokenizer_config.json b/fed-learning/non-dp/tuned-md-judge-round_5/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..bdfbeea6201781073c050947398f045f41c0531b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_5/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5052741b394891aa03f38b61fb141bd7ef264fefc70c189ec04b7a0a2964368
+size 1025
diff --git a/fed-learning/non-dp/tuned-md-judge-round_6/README.md b/fed-learning/non-dp/tuned-md-judge-round_6/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e13cffd9d2c7a10815128d520003acc68208b56f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_6/README.md
@@ -0,0 +1,202 @@
+---
+base_model: OpenSafetyLab/MD-Judge-v0.1
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-md-judge-round_6/adapter_config.json b/fed-learning/non-dp/tuned-md-judge-round_6/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..546e45178b8f16b3cdede31c83fae787b343c98b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_6/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6746a4999fdb0659f3c602f569a440da8f0cb25b35c6154630488f44f4359a7
+size 810
diff --git a/fed-learning/non-dp/tuned-md-judge-round_6/adapter_model.safetensors b/fed-learning/non-dp/tuned-md-judge-round_6/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..898705586029abde75d26c29a726edd1b663690f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_6/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebe1078b16cfa9ebc8d6b3b50849c428654067e7d0b5bc8f5e542246035a8be5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-md-judge-round_6/special_tokens_map.json b/fed-learning/non-dp/tuned-md-judge-round_6/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..1f852aa3787fa33e54d000e81d8d7004fbfe7282
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_6/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4452b845ab9ca2afa863f59f1411a959e8a32bb4334e264b0e71d5d17dd4dcb9
+size 437
diff --git a/fed-learning/non-dp/tuned-md-judge-round_6/tokenizer.json b/fed-learning/non-dp/tuned-md-judge-round_6/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..300e828335a38f4ac67a8fd713ed8c05d5c704cf
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_6/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2699839c243202a04a90537576fc719283f638e90fe80feb469888275289575
+size 3505751
diff --git a/fed-learning/non-dp/tuned-md-judge-round_6/tokenizer.model b/fed-learning/non-dp/tuned-md-judge-round_6/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..8b443ef19c2a19acc3ac64fb9c3db4a72921dff6
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_6/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443
diff --git a/fed-learning/non-dp/tuned-md-judge-round_6/tokenizer_config.json b/fed-learning/non-dp/tuned-md-judge-round_6/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..bdfbeea6201781073c050947398f045f41c0531b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_6/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5052741b394891aa03f38b61fb141bd7ef264fefc70c189ec04b7a0a2964368
+size 1025
diff --git a/fed-learning/non-dp/tuned-md-judge-round_7/README.md b/fed-learning/non-dp/tuned-md-judge-round_7/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e13cffd9d2c7a10815128d520003acc68208b56f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_7/README.md
@@ -0,0 +1,202 @@
+---
+base_model: OpenSafetyLab/MD-Judge-v0.1
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-md-judge-round_7/adapter_config.json b/fed-learning/non-dp/tuned-md-judge-round_7/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..546e45178b8f16b3cdede31c83fae787b343c98b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_7/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6746a4999fdb0659f3c602f569a440da8f0cb25b35c6154630488f44f4359a7
+size 810
diff --git a/fed-learning/non-dp/tuned-md-judge-round_7/adapter_model.safetensors b/fed-learning/non-dp/tuned-md-judge-round_7/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..898705586029abde75d26c29a726edd1b663690f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_7/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebe1078b16cfa9ebc8d6b3b50849c428654067e7d0b5bc8f5e542246035a8be5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-md-judge-round_7/special_tokens_map.json b/fed-learning/non-dp/tuned-md-judge-round_7/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..1f852aa3787fa33e54d000e81d8d7004fbfe7282
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_7/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4452b845ab9ca2afa863f59f1411a959e8a32bb4334e264b0e71d5d17dd4dcb9
+size 437
diff --git a/fed-learning/non-dp/tuned-md-judge-round_7/tokenizer.json b/fed-learning/non-dp/tuned-md-judge-round_7/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..300e828335a38f4ac67a8fd713ed8c05d5c704cf
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_7/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2699839c243202a04a90537576fc719283f638e90fe80feb469888275289575
+size 3505751
diff --git a/fed-learning/non-dp/tuned-md-judge-round_7/tokenizer.model b/fed-learning/non-dp/tuned-md-judge-round_7/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..8b443ef19c2a19acc3ac64fb9c3db4a72921dff6
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_7/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443
diff --git a/fed-learning/non-dp/tuned-md-judge-round_7/tokenizer_config.json b/fed-learning/non-dp/tuned-md-judge-round_7/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..bdfbeea6201781073c050947398f045f41c0531b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_7/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5052741b394891aa03f38b61fb141bd7ef264fefc70c189ec04b7a0a2964368
+size 1025
diff --git a/fed-learning/non-dp/tuned-md-judge-round_8/README.md b/fed-learning/non-dp/tuned-md-judge-round_8/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e13cffd9d2c7a10815128d520003acc68208b56f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_8/README.md
@@ -0,0 +1,202 @@
+---
+base_model: OpenSafetyLab/MD-Judge-v0.1
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-md-judge-round_8/adapter_config.json b/fed-learning/non-dp/tuned-md-judge-round_8/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..546e45178b8f16b3cdede31c83fae787b343c98b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_8/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6746a4999fdb0659f3c602f569a440da8f0cb25b35c6154630488f44f4359a7
+size 810
diff --git a/fed-learning/non-dp/tuned-md-judge-round_8/adapter_model.safetensors b/fed-learning/non-dp/tuned-md-judge-round_8/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..898705586029abde75d26c29a726edd1b663690f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_8/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebe1078b16cfa9ebc8d6b3b50849c428654067e7d0b5bc8f5e542246035a8be5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-md-judge-round_8/special_tokens_map.json b/fed-learning/non-dp/tuned-md-judge-round_8/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..1f852aa3787fa33e54d000e81d8d7004fbfe7282
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_8/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4452b845ab9ca2afa863f59f1411a959e8a32bb4334e264b0e71d5d17dd4dcb9
+size 437
diff --git a/fed-learning/non-dp/tuned-md-judge-round_8/tokenizer.json b/fed-learning/non-dp/tuned-md-judge-round_8/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..300e828335a38f4ac67a8fd713ed8c05d5c704cf
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_8/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2699839c243202a04a90537576fc719283f638e90fe80feb469888275289575
+size 3505751
diff --git a/fed-learning/non-dp/tuned-md-judge-round_8/tokenizer.model b/fed-learning/non-dp/tuned-md-judge-round_8/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..8b443ef19c2a19acc3ac64fb9c3db4a72921dff6
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_8/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443
diff --git a/fed-learning/non-dp/tuned-md-judge-round_8/tokenizer_config.json b/fed-learning/non-dp/tuned-md-judge-round_8/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..bdfbeea6201781073c050947398f045f41c0531b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_8/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5052741b394891aa03f38b61fb141bd7ef264fefc70c189ec04b7a0a2964368
+size 1025
diff --git a/fed-learning/non-dp/tuned-md-judge-round_9/README.md b/fed-learning/non-dp/tuned-md-judge-round_9/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e13cffd9d2c7a10815128d520003acc68208b56f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_9/README.md
@@ -0,0 +1,202 @@
+---
+base_model: OpenSafetyLab/MD-Judge-v0.1
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2
\ No newline at end of file
diff --git a/fed-learning/non-dp/tuned-md-judge-round_9/adapter_config.json b/fed-learning/non-dp/tuned-md-judge-round_9/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..546e45178b8f16b3cdede31c83fae787b343c98b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_9/adapter_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6746a4999fdb0659f3c602f569a440da8f0cb25b35c6154630488f44f4359a7
+size 810
diff --git a/fed-learning/non-dp/tuned-md-judge-round_9/adapter_model.safetensors b/fed-learning/non-dp/tuned-md-judge-round_9/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..898705586029abde75d26c29a726edd1b663690f
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_9/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebe1078b16cfa9ebc8d6b3b50849c428654067e7d0b5bc8f5e542246035a8be5
+size 27297032
diff --git a/fed-learning/non-dp/tuned-md-judge-round_9/special_tokens_map.json b/fed-learning/non-dp/tuned-md-judge-round_9/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..1f852aa3787fa33e54d000e81d8d7004fbfe7282
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_9/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4452b845ab9ca2afa863f59f1411a959e8a32bb4334e264b0e71d5d17dd4dcb9
+size 437
diff --git a/fed-learning/non-dp/tuned-md-judge-round_9/tokenizer.json b/fed-learning/non-dp/tuned-md-judge-round_9/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..300e828335a38f4ac67a8fd713ed8c05d5c704cf
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_9/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2699839c243202a04a90537576fc719283f638e90fe80feb469888275289575
+size 3505751
diff --git a/fed-learning/non-dp/tuned-md-judge-round_9/tokenizer.model b/fed-learning/non-dp/tuned-md-judge-round_9/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..8b443ef19c2a19acc3ac64fb9c3db4a72921dff6
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_9/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443
diff --git a/fed-learning/non-dp/tuned-md-judge-round_9/tokenizer_config.json b/fed-learning/non-dp/tuned-md-judge-round_9/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..bdfbeea6201781073c050947398f045f41c0531b
--- /dev/null
+++ b/fed-learning/non-dp/tuned-md-judge-round_9/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5052741b394891aa03f38b61fb141bd7ef264fefc70c189ec04b7a0a2964368
+size 1025