Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,6 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+checkpoint-100/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoint-50/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+final_checkpoint/tokenizer.json filter=lfs diff=lfs merge=lfs -text

checkpoint-100/README.md

@@ -0,0 +1,202 @@
+---
+base_model: unsloth/gemma-3-27b-it-qat-unsloth-bnb-4bit
+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

checkpoint-100/adapter_config.json

@@ -0,0 +1,39 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "unsloth/gemma-3-27b-it-qat-unsloth-bnb-4bit",
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "down_proj",
+    "o_proj",
+    "gate_proj",
+    "v_proj",
+    "up_proj",
+    "q_proj",
+    "k_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

checkpoint-100/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4dc97fecb42dcbe676206c02ddcad2cbdcad5984a51675f6a62742e3ba02e0f0
+size 932177352

checkpoint-100/added_tokens.json

@@ -0,0 +1,3 @@
+{
+  "<image_soft_token>": 262144
+}

checkpoint-100/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e03ac739c6d5d84956ff12bb97398edaffbe5f8590a68cb4d7ad791ce5178acd
+size 1817012879

checkpoint-100/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b90095a10c03d4355bd1deb440864584a708ac7ea7a9cc072a09474fe343bc3
+size 14645

checkpoint-100/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:45bb9db9b99b330f12a8646aa23c5c29fc160c9b0461608a2bc757c433d787e1
+size 1465

checkpoint-100/special_tokens_map.json

@@ -0,0 +1,33 @@
+{
+  "boi_token": "<start_of_image>",
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eoi_token": "<end_of_image>",
+  "eos_token": {
+    "content": "<end_of_turn>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "image_token": "<image_soft_token>",
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoint-100/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d786405177734910d7a3db625c2826640964a0b4e5cdbbd70620ae3313a01bef
+size 33384722

checkpoint-100/tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1299c11d7cf632ef3b4e11937501358ada021bbdf7c47638d13c0ee982f2e79c
+size 4689074

checkpoint-100/trainer_state.json

@@ -0,0 +1,1234 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.4,
+  "eval_steps": 500,
+  "global_step": 100,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "completion_length": 361.75,
+      "epoch": 0.004,
+      "grad_norm": 0.0693558007478714,
+      "kl": 0.0,
+      "learning_rate": 0.0,
+      "loss": -0.0,
+      "reward": 0.02345000021159649,
+      "reward_std": 0.04690000042319298,
+      "rewards/pot_combined_reward": 0.02345000021159649,
+      "step": 1
+    },
+    {
+      "completion_length": 371.375,
+      "epoch": 0.008,
+      "grad_norm": 0.08951307833194733,
+      "kl": 0.0,
+      "learning_rate": 5.000000000000001e-07,
+      "loss": -0.0,
+      "reward": 0.026133334264159203,
+      "reward_std": 0.052266668528318405,
+      "rewards/pot_combined_reward": 0.026133334264159203,
+      "step": 2
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.012,
+      "grad_norm": 0.001663331058807671,
+      "kl": 0.0005528016190510243,
+      "learning_rate": 1.0000000000000002e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 3
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.016,
+      "grad_norm": 0.0016158577054738998,
+      "kl": 0.0005013404006604105,
+      "learning_rate": 1.5e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 4
+    },
+    {
+      "completion_length": 373.375,
+      "epoch": 0.02,
+      "grad_norm": 0.0030017346143722534,
+      "kl": 0.0005660907772835344,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 5
+    },
+    {
+      "completion_length": 373.5625,
+      "epoch": 0.024,
+      "grad_norm": 0.0015043334569782019,
+      "kl": 0.0005426810312201269,
+      "learning_rate": 2.5e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 6
+    },
+    {
+      "completion_length": 364.5,
+      "epoch": 0.028,
+      "grad_norm": 0.08007726073265076,
+      "kl": 0.0005026786457165144,
+      "learning_rate": 3e-06,
+      "loss": 0.0001,
+      "reward": 0.03146666660904884,
+      "reward_std": 0.06293333321809769,
+      "rewards/pot_combined_reward": 0.03146666660904884,
+      "step": 7
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.032,
+      "grad_norm": 0.0016093035228550434,
+      "kl": 0.0005015511706005782,
+      "learning_rate": 3.5e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 8
+    },
+    {
+      "completion_length": 353.0625,
+      "epoch": 0.036,
+      "grad_norm": 0.001784446300007403,
+      "kl": 0.0003549655375536531,
+      "learning_rate": 4.000000000000001e-06,
+      "loss": 0.0,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 9
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.04,
+      "grad_norm": 0.0020515809301286936,
+      "kl": 0.0005762250584666617,
+      "learning_rate": 4.5e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 10
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.044,
+      "grad_norm": 0.06735244393348694,
+      "kl": 0.0005008808220736682,
+      "learning_rate": 5e-06,
+      "loss": 0.0001,
+      "reward": 0.01808333396911621,
+      "reward_std": 0.03616666793823242,
+      "rewards/pot_combined_reward": 0.01808333396911621,
+      "step": 11
+    },
+    {
+      "completion_length": 342.5,
+      "epoch": 0.048,
+      "grad_norm": 0.0024033007211983204,
+      "kl": 0.00046441886661341414,
+      "learning_rate": 4.99847706754774e-06,
+      "loss": 0.0,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 12
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.052,
+      "grad_norm": 0.0014815045287832618,
+      "kl": 0.0004604290661518462,
+      "learning_rate": 4.993910125649561e-06,
+      "loss": 0.0,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 13
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.056,
+      "grad_norm": 0.001548771746456623,
+      "kl": 0.0004989306180505082,
+      "learning_rate": 4.986304738420684e-06,
+      "loss": 0.0,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 14
+    },
+    {
+      "completion_length": 360.8125,
+      "epoch": 0.06,
+      "grad_norm": 0.001822226564399898,
+      "kl": 0.00046228048086049967,
+      "learning_rate": 4.975670171853926e-06,
+      "loss": 0.0,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 15
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.064,
+      "grad_norm": 0.09072617441415787,
+      "kl": 0.000476861278002616,
+      "learning_rate": 4.962019382530521e-06,
+      "loss": 0.0,
+      "reward": 0.012600000016391277,
+      "reward_std": 0.025200000032782555,
+      "rewards/pot_combined_reward": 0.012600000016391277,
+      "step": 16
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.068,
+      "grad_norm": 0.0015437575057148933,
+      "kl": 0.0005299622716847807,
+      "learning_rate": 4.9453690018345144e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 17
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.072,
+      "grad_norm": 0.0016528957057744265,
+      "kl": 0.0005346549514797516,
+      "learning_rate": 4.925739315689991e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 18
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.076,
+      "grad_norm": 0.0017183530144393444,
+      "kl": 0.0004841076224693097,
+      "learning_rate": 4.903154239845798e-06,
+      "loss": 0.0,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 19
+    },
+    {
+      "completion_length": 373.5625,
+      "epoch": 0.08,
+      "grad_norm": 0.08476348221302032,
+      "kl": 0.000539450986252632,
+      "learning_rate": 4.8776412907378845e-06,
+      "loss": 0.0001,
+      "reward": 0.026249999180436134,
+      "reward_std": 0.05249999836087227,
+      "rewards/pot_combined_reward": 0.026249999180436134,
+      "step": 20
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.084,
+      "grad_norm": 0.07820821553468704,
+      "kl": 0.0005957721295999363,
+      "learning_rate": 4.849231551964771e-06,
+      "loss": 0.0001,
+      "reward": 0.07116249948740005,
+      "reward_std": 0.1423249989748001,
+      "rewards/pot_combined_reward": 0.07116249948740005,
+      "step": 21
+    },
+    {
+      "completion_length": 373.5625,
+      "epoch": 0.088,
+      "grad_norm": 0.0016008545644581318,
+      "kl": 0.000533243379322812,
+      "learning_rate": 4.817959636416969e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 22
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.092,
+      "grad_norm": 0.0018651616992428899,
+      "kl": 0.0005620143201667815,
+      "learning_rate": 4.783863644106502e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 23
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.096,
+      "grad_norm": 0.001895356923341751,
+      "kl": 0.00047788477240828797,
+      "learning_rate": 4.746985115747918e-06,
+      "loss": 0.0,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 24
+    },
+    {
+      "completion_length": 369.75,
+      "epoch": 0.1,
+      "grad_norm": 0.11337540298700333,
+      "kl": 0.00047047801490407437,
+      "learning_rate": 4.707368982147318e-06,
+      "loss": 0.0,
+      "reward": 0.04736666567623615,
+      "reward_std": 0.0947333313524723,
+      "rewards/pot_combined_reward": 0.04736666567623615,
+      "step": 25
+    },
+    {
+      "completion_length": 368.625,
+      "epoch": 0.104,
+      "grad_norm": 0.01079186424612999,
+      "kl": 0.0006513141634059139,
+      "learning_rate": 4.665063509461098e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 26
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.108,
+      "grad_norm": 0.0016566955018788576,
+      "kl": 0.0005965056043351069,
+      "learning_rate": 4.620120240391065e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 27
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.112,
+      "grad_norm": 0.001792517607100308,
+      "kl": 0.0005393773099058308,
+      "learning_rate": 4.572593931387604e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 28
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.116,
+      "grad_norm": 0.0017438618233427405,
+      "kl": 0.0005095232809253503,
+      "learning_rate": 4.522542485937369e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 29
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.12,
+      "grad_norm": 0.0015684061218053102,
+      "kl": 0.00047055614413693547,
+      "learning_rate": 4.470026884016805e-06,
+      "loss": 0.0,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 30
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.124,
+      "grad_norm": 0.0019608919974416494,
+      "kl": 0.0005961552087683231,
+      "learning_rate": 4.415111107797445e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 31
+    },
+    {
+      "completion_length": 372.0625,
+      "epoch": 0.128,
+      "grad_norm": 0.0015237935585901141,
+      "kl": 0.0005326158570824191,
+      "learning_rate": 4.357862063693486e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 32
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.132,
+      "grad_norm": 0.002203061943873763,
+      "kl": 0.0006071907628211193,
+      "learning_rate": 4.2983495008466285e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 33
+    },
+    {
+      "completion_length": 370.5,
+      "epoch": 0.136,
+      "grad_norm": 0.10126212984323502,
+      "kl": 0.0006130525580374524,
+      "learning_rate": 4.236645926147493e-06,
+      "loss": 0.0001,
+      "reward": 0.061249999329447746,
+      "reward_std": 0.12249999865889549,
+      "rewards/pot_combined_reward": 0.061249999329447746,
+      "step": 34
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.14,
+      "grad_norm": 0.08621055632829666,
+      "kl": 0.0004927485424559563,
+      "learning_rate": 4.172826515897146e-06,
+      "loss": 0.0,
+      "reward": 0.024966666474938393,
+      "reward_std": 0.049933332949876785,
+      "rewards/pot_combined_reward": 0.024966666474938393,
+      "step": 35
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.144,
+      "grad_norm": 0.0017127083847299218,
+      "kl": 0.0005035524372942746,
+      "learning_rate": 4.106969024216348e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 36
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.148,
+      "grad_norm": 0.0018064226023852825,
+      "kl": 0.0005530964990612119,
+      "learning_rate": 4.039153688314146e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 37
+    },
+    {
+      "completion_length": 373.8125,
+      "epoch": 0.152,
+      "grad_norm": 0.0019408023217692971,
+      "kl": 0.0006417437689378858,
+      "learning_rate": 3.969463130731183e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 38
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.156,
+      "grad_norm": 0.08421237021684647,
+      "kl": 0.0006241849769139662,
+      "learning_rate": 3.897982258676867e-06,
+      "loss": 0.0001,
+      "reward": 0.07303333282470703,
+      "reward_std": 0.057444244623184204,
+      "rewards/pot_combined_reward": 0.07303333282470703,
+      "step": 39
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.16,
+      "grad_norm": 0.08112610131502151,
+      "kl": 0.0005679467285517603,
+      "learning_rate": 3.824798160583012e-06,
+      "loss": 0.0001,
+      "reward": 0.014233333058655262,
+      "reward_std": 0.028466666117310524,
+      "rewards/pot_combined_reward": 0.014233333058655262,
+      "step": 40
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.164,
+      "grad_norm": 0.001669050194323063,
+      "kl": 0.0005617116403300315,
+      "learning_rate": 3.7500000000000005e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 41
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.168,
+      "grad_norm": 0.002355287317186594,
+      "kl": 0.0006023006426403299,
+      "learning_rate": 3.6736789069647273e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 42
+    },
+    {
+      "completion_length": 328.8125,
+      "epoch": 0.172,
+      "grad_norm": 0.0023054229095578194,
+      "kl": 0.0005200250307098031,
+      "learning_rate": 3.595927866972694e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 43
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.176,
+      "grad_norm": 0.11641528457403183,
+      "kl": 0.0005873750924365595,
+      "learning_rate": 3.516841607689501e-06,
+      "loss": 0.0001,
+      "reward": 0.07468749955296516,
+      "reward_std": 0.14937499910593033,
+      "rewards/pot_combined_reward": 0.07468749955296516,
+      "step": 44
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.18,
+      "grad_norm": 0.001737726735882461,
+      "kl": 0.0005044558856752701,
+      "learning_rate": 3.436516483539781e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 45
+    },
+    {
+      "completion_length": 370.9375,
+      "epoch": 0.184,
+      "grad_norm": 0.0019563438836485147,
+      "kl": 0.000592117925407365,
+      "learning_rate": 3.3550503583141726e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 46
+    },
+    {
+      "completion_length": 372.4375,
+      "epoch": 0.188,
+      "grad_norm": 0.0026125519070774317,
+      "kl": 0.0005959889385849237,
+      "learning_rate": 3.272542485937369e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 47
+    },
+    {
+      "completion_length": 372.75,
+      "epoch": 0.192,
+      "grad_norm": 0.0820600688457489,
+      "kl": 0.0005973696243017912,
+      "learning_rate": 3.189093389542498e-06,
+      "loss": 0.0001,
+      "reward": 0.03968749940395355,
+      "reward_std": 0.0793749988079071,
+      "rewards/pot_combined_reward": 0.03968749940395355,
+      "step": 48
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.196,
+      "grad_norm": 0.06782057881355286,
+      "kl": 0.0007923852826934308,
+      "learning_rate": 3.1048047389991693e-06,
+      "loss": 0.0001,
+      "reward": 0.026249999180436134,
+      "reward_std": 0.05249999836087227,
+      "rewards/pot_combined_reward": 0.026249999180436134,
+      "step": 49
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.2,
+      "grad_norm": 0.10189752280712128,
+      "kl": 0.0004997247888240963,
+      "learning_rate": 3.019779227044398e-06,
+      "loss": 0.0,
+      "reward": 0.06999999843537807,
+      "reward_std": 0.13999999687075615,
+      "rewards/pot_combined_reward": 0.06999999843537807,
+      "step": 50
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.204,
+      "grad_norm": 0.004633176140487194,
+      "kl": 0.0005932025596848689,
+      "learning_rate": 2.9341204441673267e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 51
+    },
+    {
+      "completion_length": 373.3125,
+      "epoch": 0.208,
+      "grad_norm": 0.0021235239692032337,
+      "kl": 0.0006392140567186289,
+      "learning_rate": 2.847932752400164e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 52
+    },
+    {
+      "completion_length": 348.5,
+      "epoch": 0.212,
+      "grad_norm": 0.001950100064277649,
+      "kl": 0.0006046331109246239,
+      "learning_rate": 2.761321158169134e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 53
+    },
+    {
+      "completion_length": 351.875,
+      "epoch": 0.216,
+      "grad_norm": 0.009500402957201004,
+      "kl": 0.0005020819662604481,
+      "learning_rate": 2.6743911843603134e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 54
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.22,
+      "grad_norm": 0.001888828701339662,
+      "kl": 0.0005405374467954971,
+      "learning_rate": 2.587248741756253e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 55
+    },
+    {
+      "completion_length": 371.125,
+      "epoch": 0.224,
+      "grad_norm": 0.10614389926195145,
+      "kl": 0.0007851378031773493,
+      "learning_rate": 2.5e-06,
+      "loss": 0.0001,
+      "reward": 0.18312916532158852,
+      "reward_std": 0.27463946491479874,
+      "rewards/pot_combined_reward": 0.18312916532158852,
+      "step": 56
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.228,
+      "grad_norm": 0.002482097828760743,
+      "kl": 0.0006323368143057451,
+      "learning_rate": 2.4127512582437486e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 57
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.232,
+      "grad_norm": 0.0019110125722363591,
+      "kl": 0.0005738867403124459,
+      "learning_rate": 2.325608815639687e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 58
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.236,
+      "grad_norm": 0.13201673328876495,
+      "kl": 0.0008093079959508032,
+      "learning_rate": 2.238678841830867e-06,
+      "loss": 0.0001,
+      "reward": 0.09951666742563248,
+      "reward_std": 0.1495332196354866,
+      "rewards/pot_combined_reward": 0.09951666742563248,
+      "step": 59
+    },
+    {
+      "completion_length": 373.75,
+      "epoch": 0.24,
+      "grad_norm": 0.10325703024864197,
+      "kl": 0.0007794436533004045,
+      "learning_rate": 2.1520672475998374e-06,
+      "loss": 0.0001,
+      "reward": 0.0652166660875082,
+      "reward_std": 0.1304333321750164,
+      "rewards/pot_combined_reward": 0.0652166660875082,
+      "step": 60
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.244,
+      "grad_norm": 0.07717566192150116,
+      "kl": 0.0006568977987626567,
+      "learning_rate": 2.0658795558326745e-06,
+      "loss": 0.0001,
+      "reward": 0.03968749940395355,
+      "reward_std": 0.0793749988079071,
+      "rewards/pot_combined_reward": 0.03968749940395355,
+      "step": 61
+    },
+    {
+      "completion_length": 373.0625,
+      "epoch": 0.248,
+      "grad_norm": 0.0719151571393013,
+      "kl": 0.0006853110971860588,
+      "learning_rate": 1.9802207729556023e-06,
+      "loss": 0.0001,
+      "reward": 0.026249999180436134,
+      "reward_std": 0.05249999836087227,
+      "rewards/pot_combined_reward": 0.026249999180436134,
+      "step": 62
+    },
+    {
+      "completion_length": 373.625,
+      "epoch": 0.252,
+      "grad_norm": 0.07498770952224731,
+      "kl": 0.0007669057231396437,
+      "learning_rate": 1.895195261000831e-06,
+      "loss": 0.0001,
+      "reward": 0.013883333653211594,
+      "reward_std": 0.027766667306423187,
+      "rewards/pot_combined_reward": 0.013883333653211594,
+      "step": 63
+    },
+    {
+      "completion_length": 372.375,
+      "epoch": 0.256,
+      "grad_norm": 0.0024490871001034975,
+      "kl": 0.0007131954189389944,
+      "learning_rate": 1.8109066104575023e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 64
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.26,
+      "grad_norm": 0.09898767620325089,
+      "kl": 0.0008127802575472742,
+      "learning_rate": 1.7274575140626318e-06,
+      "loss": 0.0001,
+      "reward": 0.06544999964535236,
+      "reward_std": 0.09778991341590881,
+      "rewards/pot_combined_reward": 0.06544999964535236,
+      "step": 65
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.264,
+      "grad_norm": 0.08986019343137741,
+      "kl": 0.000880417152075097,
+      "learning_rate": 1.6449496416858285e-06,
+      "loss": 0.0001,
+      "reward": 0.04374999925494194,
+      "reward_std": 0.08749999850988388,
+      "rewards/pot_combined_reward": 0.04374999925494194,
+      "step": 66
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.268,
+      "grad_norm": 0.15933048725128174,
+      "kl": 0.0012064649199601263,
+      "learning_rate": 1.56348351646022e-06,
+      "loss": 0.0001,
+      "reward": 0.14419999904930592,
+      "reward_std": 0.28839999809861183,
+      "rewards/pot_combined_reward": 0.14419999904930592,
+      "step": 67
+    },
+    {
+      "completion_length": 366.875,
+      "epoch": 0.272,
+      "grad_norm": 0.0023835976608097553,
+      "kl": 0.0007049251580610871,
+      "learning_rate": 1.4831583923105e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 68
+    },
+    {
+      "completion_length": 364.9375,
+      "epoch": 0.276,
+      "grad_norm": 0.07123439759016037,
+      "kl": 0.0006189760897541419,
+      "learning_rate": 1.4040721330273063e-06,
+      "loss": 0.0001,
+      "reward": 0.08187499642372131,
+      "reward_std": 0.16374999284744263,
+      "rewards/pot_combined_reward": 0.08187499642372131,
+      "step": 69
+    },
+    {
+      "completion_length": 371.6875,
+      "epoch": 0.28,
+      "grad_norm": 0.09466277807950974,
+      "kl": 0.0009991034894483164,
+      "learning_rate": 1.3263210930352737e-06,
+      "loss": 0.0001,
+      "reward": 0.1346083376556635,
+      "reward_std": 0.2251959629356861,
+      "rewards/pot_combined_reward": 0.1346083376556635,
+      "step": 70
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.284,
+      "grad_norm": 0.09791545569896698,
+      "kl": 0.001197919569676742,
+      "learning_rate": 1.2500000000000007e-06,
+      "loss": 0.0001,
+      "reward": 0.04223333299160004,
+      "reward_std": 0.08446666970849037,
+      "rewards/pot_combined_reward": 0.04223333299160004,
+      "step": 71
+    },
+    {
+      "completion_length": 373.5,
+      "epoch": 0.288,
+      "grad_norm": 0.09683706611394882,
+      "kl": 0.001184802589705214,
+      "learning_rate": 1.1752018394169882e-06,
+      "loss": 0.0001,
+      "reward": 0.05285000056028366,
+      "reward_std": 0.10570000112056732,
+      "rewards/pot_combined_reward": 0.05285000056028366,
+      "step": 72
+    },
+    {
+      "completion_length": 367.25,
+      "epoch": 0.292,
+      "grad_norm": 0.0033558327704668045,
+      "kl": 0.0008135271054925397,
+      "learning_rate": 1.1020177413231334e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 73
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.296,
+      "grad_norm": 0.07025693356990814,
+      "kl": 0.0007950128638185561,
+      "learning_rate": 1.0305368692688175e-06,
+      "loss": 0.0001,
+      "reward": 0.04374999925494194,
+      "reward_std": 0.08749999850988388,
+      "rewards/pot_combined_reward": 0.04374999925494194,
+      "step": 74
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.3,
+      "grad_norm": 0.07400605827569962,
+      "kl": 0.0009908084612106904,
+      "learning_rate": 9.608463116858544e-07,
+      "loss": 0.0001,
+      "reward": 0.04339999705553055,
+      "reward_std": 0.05020662397146225,
+      "rewards/pot_combined_reward": 0.04339999705553055,
+      "step": 75
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.304,
+      "grad_norm": 0.09700194746255875,
+      "kl": 0.0008199879812309518,
+      "learning_rate": 8.930309757836517e-07,
+      "loss": 0.0001,
+      "reward": 0.061249999329447746,
+      "reward_std": 0.12249999865889549,
+      "rewards/pot_combined_reward": 0.061249999329447746,
+      "step": 76
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.308,
+      "grad_norm": 0.11501616984605789,
+      "kl": 0.001101230038329959,
+      "learning_rate": 8.271734841028553e-07,
+      "loss": 0.0001,
+      "reward": 0.12562499567866325,
+      "reward_std": 0.2512499913573265,
+      "rewards/pot_combined_reward": 0.12562499567866325,
+      "step": 77
+    },
+    {
+      "completion_length": 369.5,
+      "epoch": 0.312,
+      "grad_norm": 0.07562494277954102,
+      "kl": 0.0013928793196100742,
+      "learning_rate": 7.633540738525066e-07,
+      "loss": 0.0001,
+      "reward": 0.021816667169332504,
+      "reward_std": 0.04363333433866501,
+      "rewards/pot_combined_reward": 0.021816667169332504,
+      "step": 78
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.316,
+      "grad_norm": 0.0032410642597824335,
+      "kl": 0.0008720912301214412,
+      "learning_rate": 7.016504991533727e-07,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 79
+    },
+    {
+      "completion_length": 331.1875,
+      "epoch": 0.32,
+      "grad_norm": 0.004254704806953669,
+      "kl": 0.0010208465537289158,
+      "learning_rate": 6.421379363065142e-07,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 80
+    },
+    {
+      "completion_length": 373.75,
+      "epoch": 0.324,
+      "grad_norm": 0.002906290115788579,
+      "kl": 0.0007839756435714662,
+      "learning_rate": 5.848888922025553e-07,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 81
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.328,
+      "grad_norm": 0.13924577832221985,
+      "kl": 0.001414620659488719,
+      "learning_rate": 5.299731159831953e-07,
+      "loss": 0.0001,
+      "reward": 0.15825624577701092,
+      "reward_std": 0.31651249155402184,
+      "rewards/pot_combined_reward": 0.15825624577701092,
+      "step": 82
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.332,
+      "grad_norm": 0.14953076839447021,
+      "kl": 0.0011789420095738024,
+      "learning_rate": 4.774575140626317e-07,
+      "loss": 0.0001,
+      "reward": 0.1184374988079071,
+      "reward_std": 0.2368749976158142,
+      "rewards/pot_combined_reward": 0.1184374988079071,
+      "step": 83
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.336,
+      "grad_norm": 0.16725599765777588,
+      "kl": 0.0009174157894449309,
+      "learning_rate": 4.27406068612396e-07,
+      "loss": 0.0001,
+      "reward": 0.10500000044703484,
+      "reward_std": 0.21000000089406967,
+      "rewards/pot_combined_reward": 0.10500000044703484,
+      "step": 84
+    },
+    {
+      "completion_length": 370.6875,
+      "epoch": 0.34,
+      "grad_norm": 0.08203138411045074,
+      "kl": 0.0009189353149849921,
+      "learning_rate": 3.798797596089351e-07,
+      "loss": 0.0001,
+      "reward": 0.04374999925494194,
+      "reward_std": 0.08749999850988388,
+      "rewards/pot_combined_reward": 0.04374999925494194,
+      "step": 85
+    },
+    {
+      "completion_length": 371.5625,
+      "epoch": 0.344,
+      "grad_norm": 0.07274218648672104,
+      "kl": 0.0008360931678907946,
+      "learning_rate": 3.3493649053890325e-07,
+      "loss": 0.0001,
+      "reward": 0.08656249940395355,
+      "reward_std": 0.1731249988079071,
+      "rewards/pot_combined_reward": 0.08656249940395355,
+      "step": 86
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.348,
+      "grad_norm": 0.12336569279432297,
+      "kl": 0.0012354136852081865,
+      "learning_rate": 2.9263101785268253e-07,
+      "loss": 0.0001,
+      "reward": 0.09249166399240494,
+      "reward_std": 0.18498332425951958,
+      "rewards/pot_combined_reward": 0.09249166399240494,
+      "step": 87
+    },
+    {
+      "completion_length": 367.875,
+      "epoch": 0.352,
+      "grad_norm": 0.13889098167419434,
+      "kl": 0.0017657574207987636,
+      "learning_rate": 2.53014884252083e-07,
+      "loss": 0.0002,
+      "reward": 0.2926562614738941,
+      "reward_std": 0.4413819834589958,
+      "rewards/pot_combined_reward": 0.2926562614738941,
+      "step": 88
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.356,
+      "grad_norm": 0.06699243932962418,
+      "kl": 0.001279524396522902,
+      "learning_rate": 2.1613635589349756e-07,
+      "loss": 0.0001,
+      "reward": 0.08187499642372131,
+      "reward_std": 0.16374999284744263,
+      "rewards/pot_combined_reward": 0.08187499642372131,
+      "step": 89
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.36,
+      "grad_norm": 0.006427991669625044,
+      "kl": 0.0013257948303362355,
+      "learning_rate": 1.8204036358303173e-07,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 90
+    },
+    {
+      "completion_length": 370.9375,
+      "epoch": 0.364,
+      "grad_norm": 0.1104293167591095,
+      "kl": 0.0018631000712048262,
+      "learning_rate": 1.507684480352292e-07,
+      "loss": 0.0002,
+      "reward": 0.058799998834729195,
+      "reward_std": 0.11759999766945839,
+      "rewards/pot_combined_reward": 0.058799998834729195,
+      "step": 91
+    },
+    {
+      "completion_length": 361.9375,
+      "epoch": 0.368,
+      "grad_norm": 0.07179438322782516,
+      "kl": 0.0009191952631226741,
+      "learning_rate": 1.223587092621162e-07,
+      "loss": 0.0001,
+      "reward": 0.08343750238418579,
+      "reward_std": 0.09881261736154556,
+      "rewards/pot_combined_reward": 0.08343750238418579,
+      "step": 92
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.372,
+      "grad_norm": 0.06493149697780609,
+      "kl": 0.0012784109567292035,
+      "learning_rate": 9.684576015420277e-08,
+      "loss": 0.0001,
+      "reward": 0.11232500523328781,
+      "reward_std": 0.17172931134700775,
+      "rewards/pot_combined_reward": 0.11232500523328781,
+      "step": 93
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.376,
+      "grad_norm": 0.12131404131650925,
+      "kl": 0.0012257406779099256,
+      "learning_rate": 7.426068431000883e-08,
+      "loss": 0.0001,
+      "reward": 0.1303124949336052,
+      "reward_std": 0.2606249898672104,
+      "rewards/pot_combined_reward": 0.1303124949336052,
+      "step": 94
+    },
+    {
+      "completion_length": 368.0625,
+      "epoch": 0.38,
+      "grad_norm": 0.1013365238904953,
+      "kl": 0.0013991060986882076,
+      "learning_rate": 5.463099816548578e-08,
+      "loss": 0.0001,
+      "reward": 0.10476666688919067,
+      "reward_std": 0.15946697443723679,
+      "rewards/pot_combined_reward": 0.10476666688919067,
+      "step": 95
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.384,
+      "grad_norm": 0.06952964514493942,
+      "kl": 0.0014077925588935614,
+      "learning_rate": 3.798061746947995e-08,
+      "loss": 0.0001,
+      "reward": 0.11687499284744263,
+      "reward_std": 0.1551528126001358,
+      "rewards/pot_combined_reward": 0.11687499284744263,
+      "step": 96
+    },
+    {
+      "completion_length": 370.25,
+      "epoch": 0.388,
+      "grad_norm": 0.09735430032014847,
+      "kl": 0.0009093381959246472,
+      "learning_rate": 2.4329828146074096e-08,
+      "loss": 0.0001,
+      "reward": 0.026249999180436134,
+      "reward_std": 0.05249999836087227,
+      "rewards/pot_combined_reward": 0.026249999180436134,
+      "step": 97
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.392,
+      "grad_norm": 0.14300031960010529,
+      "kl": 0.0015978575684130192,
+      "learning_rate": 1.3695261579316776e-08,
+      "loss": 0.0002,
+      "reward": 0.23843749426305294,
+      "reward_std": 0.33464543148875237,
+      "rewards/pot_combined_reward": 0.23843749426305294,
+      "step": 98
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.396,
+      "grad_norm": 0.13475055992603302,
+      "kl": 0.0017390053835697472,
+      "learning_rate": 6.089874350439507e-09,
+      "loss": 0.0002,
+      "reward": 0.15921874903142452,
+      "reward_std": 0.31843749806284904,
+      "rewards/pot_combined_reward": 0.15921874903142452,
+      "step": 99
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.4,
+      "grad_norm": 0.07442972809076309,
+      "kl": 0.0010832307452801615,
+      "learning_rate": 1.5229324522605949e-09,
+      "loss": 0.0001,
+      "reward": 0.019833333790302277,
+      "reward_std": 0.03966666758060455,
+      "rewards/pot_combined_reward": 0.019833333790302277,
+      "step": 100
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 100,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 50,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 0.0,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-100/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:95f6fbb4bc99ccc7d1236fac55bcaad60028a6bad1c2bc7cdf9af60d934d017c
+size 6225

checkpoint-50/README.md

@@ -0,0 +1,202 @@
+---
+base_model: unsloth/gemma-3-27b-it-qat-unsloth-bnb-4bit
+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

checkpoint-50/adapter_config.json

@@ -0,0 +1,39 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "unsloth/gemma-3-27b-it-qat-unsloth-bnb-4bit",
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "down_proj",
+    "o_proj",
+    "gate_proj",
+    "v_proj",
+    "up_proj",
+    "q_proj",
+    "k_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

checkpoint-50/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a060b33eba5705aa23812dc5eec0de729a1d86c5be1ed92fedcee9f2f8a2c0d8
+size 932177352

checkpoint-50/added_tokens.json

@@ -0,0 +1,3 @@
+{
+  "<image_soft_token>": 262144
+}

checkpoint-50/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6fb8843aa3651828d1e6719194d09b3222738050fd6b8c802a8a281a6dcf0c43
+size 1817012879

checkpoint-50/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d4b77d6789ca4e0972492ad819454e59842839b43004c79ec142dde757de90b0
+size 14645

checkpoint-50/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b15b36a4919f1f382ee53c1ef0ce3da84250e386c000bcb8d6a58c56ab1973dd
+size 1465

checkpoint-50/special_tokens_map.json

@@ -0,0 +1,33 @@
+{
+  "boi_token": "<start_of_image>",
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eoi_token": "<end_of_image>",
+  "eos_token": {
+    "content": "<end_of_turn>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "image_token": "<image_soft_token>",
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoint-50/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d786405177734910d7a3db625c2826640964a0b4e5cdbbd70620ae3313a01bef
+size 33384722

checkpoint-50/tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1299c11d7cf632ef3b4e11937501358ada021bbdf7c47638d13c0ee982f2e79c
+size 4689074

checkpoint-50/trainer_state.json

@@ -0,0 +1,634 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.2,
+  "eval_steps": 500,
+  "global_step": 50,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "completion_length": 361.75,
+      "epoch": 0.004,
+      "grad_norm": 0.0693558007478714,
+      "kl": 0.0,
+      "learning_rate": 0.0,
+      "loss": -0.0,
+      "reward": 0.02345000021159649,
+      "reward_std": 0.04690000042319298,
+      "rewards/pot_combined_reward": 0.02345000021159649,
+      "step": 1
+    },
+    {
+      "completion_length": 371.375,
+      "epoch": 0.008,
+      "grad_norm": 0.08951307833194733,
+      "kl": 0.0,
+      "learning_rate": 5.000000000000001e-07,
+      "loss": -0.0,
+      "reward": 0.026133334264159203,
+      "reward_std": 0.052266668528318405,
+      "rewards/pot_combined_reward": 0.026133334264159203,
+      "step": 2
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.012,
+      "grad_norm": 0.001663331058807671,
+      "kl": 0.0005528016190510243,
+      "learning_rate": 1.0000000000000002e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 3
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.016,
+      "grad_norm": 0.0016158577054738998,
+      "kl": 0.0005013404006604105,
+      "learning_rate": 1.5e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 4
+    },
+    {
+      "completion_length": 373.375,
+      "epoch": 0.02,
+      "grad_norm": 0.0030017346143722534,
+      "kl": 0.0005660907772835344,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 5
+    },
+    {
+      "completion_length": 373.5625,
+      "epoch": 0.024,
+      "grad_norm": 0.0015043334569782019,
+      "kl": 0.0005426810312201269,
+      "learning_rate": 2.5e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 6
+    },
+    {
+      "completion_length": 364.5,
+      "epoch": 0.028,
+      "grad_norm": 0.08007726073265076,
+      "kl": 0.0005026786457165144,
+      "learning_rate": 3e-06,
+      "loss": 0.0001,
+      "reward": 0.03146666660904884,
+      "reward_std": 0.06293333321809769,
+      "rewards/pot_combined_reward": 0.03146666660904884,
+      "step": 7
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.032,
+      "grad_norm": 0.0016093035228550434,
+      "kl": 0.0005015511706005782,
+      "learning_rate": 3.5e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 8
+    },
+    {
+      "completion_length": 353.0625,
+      "epoch": 0.036,
+      "grad_norm": 0.001784446300007403,
+      "kl": 0.0003549655375536531,
+      "learning_rate": 4.000000000000001e-06,
+      "loss": 0.0,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 9
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.04,
+      "grad_norm": 0.0020515809301286936,
+      "kl": 0.0005762250584666617,
+      "learning_rate": 4.5e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 10
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.044,
+      "grad_norm": 0.06735244393348694,
+      "kl": 0.0005008808220736682,
+      "learning_rate": 5e-06,
+      "loss": 0.0001,
+      "reward": 0.01808333396911621,
+      "reward_std": 0.03616666793823242,
+      "rewards/pot_combined_reward": 0.01808333396911621,
+      "step": 11
+    },
+    {
+      "completion_length": 342.5,
+      "epoch": 0.048,
+      "grad_norm": 0.0024033007211983204,
+      "kl": 0.00046441886661341414,
+      "learning_rate": 4.99847706754774e-06,
+      "loss": 0.0,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 12
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.052,
+      "grad_norm": 0.0014815045287832618,
+      "kl": 0.0004604290661518462,
+      "learning_rate": 4.993910125649561e-06,
+      "loss": 0.0,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 13
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.056,
+      "grad_norm": 0.001548771746456623,
+      "kl": 0.0004989306180505082,
+      "learning_rate": 4.986304738420684e-06,
+      "loss": 0.0,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 14
+    },
+    {
+      "completion_length": 360.8125,
+      "epoch": 0.06,
+      "grad_norm": 0.001822226564399898,
+      "kl": 0.00046228048086049967,
+      "learning_rate": 4.975670171853926e-06,
+      "loss": 0.0,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 15
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.064,
+      "grad_norm": 0.09072617441415787,
+      "kl": 0.000476861278002616,
+      "learning_rate": 4.962019382530521e-06,
+      "loss": 0.0,
+      "reward": 0.012600000016391277,
+      "reward_std": 0.025200000032782555,
+      "rewards/pot_combined_reward": 0.012600000016391277,
+      "step": 16
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.068,
+      "grad_norm": 0.0015437575057148933,
+      "kl": 0.0005299622716847807,
+      "learning_rate": 4.9453690018345144e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 17
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.072,
+      "grad_norm": 0.0016528957057744265,
+      "kl": 0.0005346549514797516,
+      "learning_rate": 4.925739315689991e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 18
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.076,
+      "grad_norm": 0.0017183530144393444,
+      "kl": 0.0004841076224693097,
+      "learning_rate": 4.903154239845798e-06,
+      "loss": 0.0,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 19
+    },
+    {
+      "completion_length": 373.5625,
+      "epoch": 0.08,
+      "grad_norm": 0.08476348221302032,
+      "kl": 0.000539450986252632,
+      "learning_rate": 4.8776412907378845e-06,
+      "loss": 0.0001,
+      "reward": 0.026249999180436134,
+      "reward_std": 0.05249999836087227,
+      "rewards/pot_combined_reward": 0.026249999180436134,
+      "step": 20
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.084,
+      "grad_norm": 0.07820821553468704,
+      "kl": 0.0005957721295999363,
+      "learning_rate": 4.849231551964771e-06,
+      "loss": 0.0001,
+      "reward": 0.07116249948740005,
+      "reward_std": 0.1423249989748001,
+      "rewards/pot_combined_reward": 0.07116249948740005,
+      "step": 21
+    },
+    {
+      "completion_length": 373.5625,
+      "epoch": 0.088,
+      "grad_norm": 0.0016008545644581318,
+      "kl": 0.000533243379322812,
+      "learning_rate": 4.817959636416969e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 22
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.092,
+      "grad_norm": 0.0018651616992428899,
+      "kl": 0.0005620143201667815,
+      "learning_rate": 4.783863644106502e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 23
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.096,
+      "grad_norm": 0.001895356923341751,
+      "kl": 0.00047788477240828797,
+      "learning_rate": 4.746985115747918e-06,
+      "loss": 0.0,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 24
+    },
+    {
+      "completion_length": 369.75,
+      "epoch": 0.1,
+      "grad_norm": 0.11337540298700333,
+      "kl": 0.00047047801490407437,
+      "learning_rate": 4.707368982147318e-06,
+      "loss": 0.0,
+      "reward": 0.04736666567623615,
+      "reward_std": 0.0947333313524723,
+      "rewards/pot_combined_reward": 0.04736666567623615,
+      "step": 25
+    },
+    {
+      "completion_length": 368.625,
+      "epoch": 0.104,
+      "grad_norm": 0.01079186424612999,
+      "kl": 0.0006513141634059139,
+      "learning_rate": 4.665063509461098e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 26
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.108,
+      "grad_norm": 0.0016566955018788576,
+      "kl": 0.0005965056043351069,
+      "learning_rate": 4.620120240391065e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 27
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.112,
+      "grad_norm": 0.001792517607100308,
+      "kl": 0.0005393773099058308,
+      "learning_rate": 4.572593931387604e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 28
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.116,
+      "grad_norm": 0.0017438618233427405,
+      "kl": 0.0005095232809253503,
+      "learning_rate": 4.522542485937369e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 29
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.12,
+      "grad_norm": 0.0015684061218053102,
+      "kl": 0.00047055614413693547,
+      "learning_rate": 4.470026884016805e-06,
+      "loss": 0.0,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 30
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.124,
+      "grad_norm": 0.0019608919974416494,
+      "kl": 0.0005961552087683231,
+      "learning_rate": 4.415111107797445e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 31
+    },
+    {
+      "completion_length": 372.0625,
+      "epoch": 0.128,
+      "grad_norm": 0.0015237935585901141,
+      "kl": 0.0005326158570824191,
+      "learning_rate": 4.357862063693486e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 32
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.132,
+      "grad_norm": 0.002203061943873763,
+      "kl": 0.0006071907628211193,
+      "learning_rate": 4.2983495008466285e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 33
+    },
+    {
+      "completion_length": 370.5,
+      "epoch": 0.136,
+      "grad_norm": 0.10126212984323502,
+      "kl": 0.0006130525580374524,
+      "learning_rate": 4.236645926147493e-06,
+      "loss": 0.0001,
+      "reward": 0.061249999329447746,
+      "reward_std": 0.12249999865889549,
+      "rewards/pot_combined_reward": 0.061249999329447746,
+      "step": 34
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.14,
+      "grad_norm": 0.08621055632829666,
+      "kl": 0.0004927485424559563,
+      "learning_rate": 4.172826515897146e-06,
+      "loss": 0.0,
+      "reward": 0.024966666474938393,
+      "reward_std": 0.049933332949876785,
+      "rewards/pot_combined_reward": 0.024966666474938393,
+      "step": 35
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.144,
+      "grad_norm": 0.0017127083847299218,
+      "kl": 0.0005035524372942746,
+      "learning_rate": 4.106969024216348e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 36
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.148,
+      "grad_norm": 0.0018064226023852825,
+      "kl": 0.0005530964990612119,
+      "learning_rate": 4.039153688314146e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 37
+    },
+    {
+      "completion_length": 373.8125,
+      "epoch": 0.152,
+      "grad_norm": 0.0019408023217692971,
+      "kl": 0.0006417437689378858,
+      "learning_rate": 3.969463130731183e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 38
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.156,
+      "grad_norm": 0.08421237021684647,
+      "kl": 0.0006241849769139662,
+      "learning_rate": 3.897982258676867e-06,
+      "loss": 0.0001,
+      "reward": 0.07303333282470703,
+      "reward_std": 0.057444244623184204,
+      "rewards/pot_combined_reward": 0.07303333282470703,
+      "step": 39
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.16,
+      "grad_norm": 0.08112610131502151,
+      "kl": 0.0005679467285517603,
+      "learning_rate": 3.824798160583012e-06,
+      "loss": 0.0001,
+      "reward": 0.014233333058655262,
+      "reward_std": 0.028466666117310524,
+      "rewards/pot_combined_reward": 0.014233333058655262,
+      "step": 40
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.164,
+      "grad_norm": 0.001669050194323063,
+      "kl": 0.0005617116403300315,
+      "learning_rate": 3.7500000000000005e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 41
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.168,
+      "grad_norm": 0.002355287317186594,
+      "kl": 0.0006023006426403299,
+      "learning_rate": 3.6736789069647273e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 42
+    },
+    {
+      "completion_length": 328.8125,
+      "epoch": 0.172,
+      "grad_norm": 0.0023054229095578194,
+      "kl": 0.0005200250307098031,
+      "learning_rate": 3.595927866972694e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 43
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.176,
+      "grad_norm": 0.11641528457403183,
+      "kl": 0.0005873750924365595,
+      "learning_rate": 3.516841607689501e-06,
+      "loss": 0.0001,
+      "reward": 0.07468749955296516,
+      "reward_std": 0.14937499910593033,
+      "rewards/pot_combined_reward": 0.07468749955296516,
+      "step": 44
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.18,
+      "grad_norm": 0.001737726735882461,
+      "kl": 0.0005044558856752701,
+      "learning_rate": 3.436516483539781e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 45
+    },
+    {
+      "completion_length": 370.9375,
+      "epoch": 0.184,
+      "grad_norm": 0.0019563438836485147,
+      "kl": 0.000592117925407365,
+      "learning_rate": 3.3550503583141726e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 46
+    },
+    {
+      "completion_length": 372.4375,
+      "epoch": 0.188,
+      "grad_norm": 0.0026125519070774317,
+      "kl": 0.0005959889385849237,
+      "learning_rate": 3.272542485937369e-06,
+      "loss": 0.0001,
+      "reward": 0.0,
+      "reward_std": 0.0,
+      "rewards/pot_combined_reward": 0.0,
+      "step": 47
+    },
+    {
+      "completion_length": 372.75,
+      "epoch": 0.192,
+      "grad_norm": 0.0820600688457489,
+      "kl": 0.0005973696243017912,
+      "learning_rate": 3.189093389542498e-06,
+      "loss": 0.0001,
+      "reward": 0.03968749940395355,
+      "reward_std": 0.0793749988079071,
+      "rewards/pot_combined_reward": 0.03968749940395355,
+      "step": 48
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.196,
+      "grad_norm": 0.06782057881355286,
+      "kl": 0.0007923852826934308,
+      "learning_rate": 3.1048047389991693e-06,
+      "loss": 0.0001,
+      "reward": 0.026249999180436134,
+      "reward_std": 0.05249999836087227,
+      "rewards/pot_combined_reward": 0.026249999180436134,
+      "step": 49
+    },
+    {
+      "completion_length": 374.0,
+      "epoch": 0.2,
+      "grad_norm": 0.10189752280712128,
+      "kl": 0.0004997247888240963,
+      "learning_rate": 3.019779227044398e-06,
+      "loss": 0.0,
+      "reward": 0.06999999843537807,
+      "reward_std": 0.13999999687075615,
+      "rewards/pot_combined_reward": 0.06999999843537807,
+      "step": 50
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 100,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 50,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 0.0,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-50/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:95f6fbb4bc99ccc7d1236fac55bcaad60028a6bad1c2bc7cdf9af60d934d017c
+size 6225

final_checkpoint/README.md

@@ -0,0 +1,202 @@
+---
+base_model: unsloth/gemma-3-27b-it-qat-unsloth-bnb-4bit
+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

final_checkpoint/adapter_config.json

@@ -0,0 +1,39 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "unsloth/gemma-3-27b-it-qat-unsloth-bnb-4bit",
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "down_proj",
+    "o_proj",
+    "gate_proj",
+    "v_proj",
+    "up_proj",
+    "q_proj",
+    "k_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

final_checkpoint/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4dc97fecb42dcbe676206c02ddcad2cbdcad5984a51675f6a62742e3ba02e0f0
+size 932177352

final_checkpoint/added_tokens.json

@@ -0,0 +1,3 @@
+{
+  "<image_soft_token>": 262144
+}

final_checkpoint/special_tokens_map.json

@@ -0,0 +1,33 @@
+{
+  "boi_token": "<start_of_image>",
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eoi_token": "<end_of_image>",
+  "eos_token": {
+    "content": "<end_of_turn>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "image_token": "<image_soft_token>",
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

final_checkpoint/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d786405177734910d7a3db625c2826640964a0b4e5cdbbd70620ae3313a01bef
+size 33384722

final_checkpoint/tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1299c11d7cf632ef3b4e11937501358ada021bbdf7c47638d13c0ee982f2e79c
+size 4689074

final_checkpoint/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:95f6fbb4bc99ccc7d1236fac55bcaad60028a6bad1c2bc7cdf9af60d934d017c
+size 6225