Model save

README.md

@@ -1,5 +1,5 @@
 ---
-base_model: meta-llama/Llama-3.2-3B-Instruct
+base_model: meta-llama/Llama-3.2-1B-Instruct
 library_name: peft
 license: llama3.2
 tags:
@@ -14,7 +14,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # lora
 
-This model is a fine-tuned version of [meta-llama/Llama-3.2-3B-Instruct](https://huggingface.co/meta-llama/Llama-3.2-3B-Instruct) on an unknown dataset.
+This model is a fine-tuned version of [meta-llama/Llama-3.2-1B-Instruct](https://huggingface.co/meta-llama/Llama-3.2-1B-Instruct) on an unknown dataset.
 
 ## Model description
 
@@ -42,7 +42,7 @@ The following hyperparameters were used during training:
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 2
-- num_epochs: 2
+- num_epochs: 4
 
 ### Training results
 
@@ -50,7 +50,7 @@ The following hyperparameters were used during training:
 
 ### Framework versions
 
-- PEFT 0.13.1
+- PEFT 0.13.0
 - Transformers 4.45.2
-- Pytorch 2.4.1
+- Pytorch 2.4.0
 - Tokenizers 0.20.0

adapter_config.json

@@ -4,7 +4,7 @@
     "base_model_class": "LlamaForCausalLM",
     "parent_library": "transformers.models.llama.modeling_llama"
   },
-  "base_model_name_or_path": "meta-llama/Llama-3.2-3B-Instruct",
+  "base_model_name_or_path": "meta-llama/Llama-3.2-1B-Instruct",
   "bias": "none",
   "fan_in_fan_out": false,
   "inference_mode": true,
@@ -24,12 +24,12 @@
   "revision": null,
   "target_modules": [
     "o_proj",
-    "k_proj",
+    "gate_proj",
     "down_proj",
     "q_proj",
-    "v_proj",
+    "k_proj",
     "up_proj",
-    "gate_proj"
+    "v_proj"
   ],
   "task_type": null,
   "use_dora": true,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:edc35461591e376a149bc97ee20d05ba3e08b63b4d1f0a96c620390206290710
-size 1639093264
+oid sha256:4362f75679bf4f5e357de009c0daa6baf45eca037300bc533ce7c7af7172c500
+size 1082290000

trainer_state.json

@@ -1,55 +1,83 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 2.0,
+  "epoch": 4.0,
   "eval_steps": 500,
-  "global_step": 1030,
+  "global_step": 2060,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.4854368932038835,
-      "grad_norm": 2.154229164123535,
-      "learning_rate": 7.587548638132295e-05,
-      "loss": 0.6104,
+      "grad_norm": 2.1187877655029297,
+      "learning_rate": 8.794946550048592e-05,
+      "loss": 0.6386,
       "step": 250
     },
     {
       "epoch": 0.970873786407767,
-      "grad_norm": 1.9545427560806274,
-      "learning_rate": 5.155642023346303e-05,
-      "loss": 0.3677,
+      "grad_norm": 1.740912675857544,
+      "learning_rate": 7.580174927113704e-05,
+      "loss": 0.3835,
       "step": 500
     },
     {
       "epoch": 1.4563106796116505,
-      "grad_norm": 2.771542549133301,
-      "learning_rate": 2.7237354085603113e-05,
-      "loss": 0.3307,
+      "grad_norm": 1.9458281993865967,
+      "learning_rate": 6.365403304178815e-05,
+      "loss": 0.3585,
       "step": 750
     },
     {
       "epoch": 1.941747572815534,
-      "grad_norm": 2.780686140060425,
-      "learning_rate": 2.9182879377431908e-06,
-      "loss": 0.2978,
+      "grad_norm": 2.2465028762817383,
+      "learning_rate": 5.150631681243926e-05,
+      "loss": 0.3352,
       "step": 1000
     },
     {
-      "epoch": 2.0,
-      "step": 1030,
-      "total_flos": 7.232215420241879e+17,
-      "train_loss": 0.39821495037634397,
-      "train_runtime": 29524.5328,
-      "train_samples_per_second": 8.931,
-      "train_steps_per_second": 0.035
+      "epoch": 2.4271844660194173,
+      "grad_norm": 2.5679686069488525,
+      "learning_rate": 3.9358600583090386e-05,
+      "loss": 0.3078,
+      "step": 1250
+    },
+    {
+      "epoch": 2.912621359223301,
+      "grad_norm": 2.835810899734497,
+      "learning_rate": 2.72108843537415e-05,
+      "loss": 0.2888,
+      "step": 1500
+    },
+    {
+      "epoch": 3.3980582524271843,
+      "grad_norm": 2.606696367263794,
+      "learning_rate": 1.5063168124392615e-05,
+      "loss": 0.2672,
+      "step": 1750
+    },
+    {
+      "epoch": 3.883495145631068,
+      "grad_norm": 2.562168836593628,
+      "learning_rate": 2.915451895043732e-06,
+      "loss": 0.2569,
+      "step": 2000
+    },
+    {
+      "epoch": 4.0,
+      "step": 2060,
+      "total_flos": 4.998764563388621e+17,
+      "train_loss": 0.35164398545200387,
+      "train_runtime": 15327.6354,
+      "train_samples_per_second": 34.405,
+      "train_steps_per_second": 0.134
     }
   ],
   "logging_steps": 250,
-  "max_steps": 1030,
+  "max_steps": 2060,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 2,
+  "num_train_epochs": 4,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -63,7 +91,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 7.232215420241879e+17,
+  "total_flos": 4.998764563388621e+17,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:04956adb5217175aaa36e11b2da87eace1a9f30a2a1627cc233a8fb4044febf4
+oid sha256:8633d9c0f297b452fd7e6997b1a857978104ecf0f156a35e33ed6acc4cb7b486
 size 5176