End of training

README.md

@@ -4,6 +4,7 @@ license: apache-2.0
 base_model: mistralai/Mistral-7B-v0.1
 tags:
 - llama-factory
+- full
 - generated_from_trainer
 model-index:
 - name: hp_ablations_grid_mistral_bsz4096_lr2e-6_scheduler-cosine-warmup0.15-minlr5e-7
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # hp_ablations_grid_mistral_bsz4096_lr2e-6_scheduler-cosine-warmup0.15-minlr5e-7
 
-This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on an unknown dataset.
+This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the mlfoundations-dev/oh-dcft-v3-llama3.1-nemotron-70b_shareGPT_format dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.0628
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 3.0,
+    "eval_loss": 0.06279128044843674,
+    "eval_runtime": 88.6978,
+    "eval_samples_per_second": 204.097,
+    "eval_steps_per_second": 0.406,
+    "total_flos": 3375200049561600.0,
+    "train_loss": 0.5529132849640317,
+    "train_runtime": 15188.2312,
+    "train_samples_per_second": 67.938,
+    "train_steps_per_second": 0.017
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 3.0,
+    "eval_loss": 0.06279128044843674,
+    "eval_runtime": 88.6978,
+    "eval_samples_per_second": 204.097,
+    "eval_steps_per_second": 0.406
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "total_flos": 3375200049561600.0,
+    "train_loss": 0.5529132849640317,
+    "train_runtime": 15188.2312,
+    "train_samples_per_second": 67.938,
+    "train_steps_per_second": 0.017
+}

trainer_state.json

@@ -0,0 +1,241 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.0,
+  "eval_steps": 500,
+  "global_step": 252,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.11904761904761904,
+      "grad_norm": 8.735864265285194,
+      "learning_rate": 5.263157894736842e-07,
+      "loss": 0.8721,
+      "step": 10
+    },
+    {
+      "epoch": 0.23809523809523808,
+      "grad_norm": 3.0586867449133153,
+      "learning_rate": 1.0526315789473683e-06,
+      "loss": 0.7667,
+      "step": 20
+    },
+    {
+      "epoch": 0.35714285714285715,
+      "grad_norm": 1.3919491548026783,
+      "learning_rate": 1.5789473684210526e-06,
+      "loss": 0.6797,
+      "step": 30
+    },
+    {
+      "epoch": 0.47619047619047616,
+      "grad_norm": 1.3015106207743352,
+      "learning_rate": 1.9996767546702485e-06,
+      "loss": 0.6273,
+      "step": 40
+    },
+    {
+      "epoch": 0.5952380952380952,
+      "grad_norm": 3.14961074638027,
+      "learning_rate": 1.988392397752233e-06,
+      "loss": 0.5986,
+      "step": 50
+    },
+    {
+      "epoch": 0.7142857142857143,
+      "grad_norm": 2.02851221384901,
+      "learning_rate": 1.961223330122206e-06,
+      "loss": 0.581,
+      "step": 60
+    },
+    {
+      "epoch": 0.8333333333333334,
+      "grad_norm": 2.523772811799601,
+      "learning_rate": 1.9187540279759314e-06,
+      "loss": 0.5693,
+      "step": 70
+    },
+    {
+      "epoch": 0.9523809523809523,
+      "grad_norm": 1.7965113435726119,
+      "learning_rate": 1.861898114721218e-06,
+      "loss": 0.5579,
+      "step": 80
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.06908344477415085,
+      "eval_runtime": 89.6395,
+      "eval_samples_per_second": 201.953,
+      "eval_steps_per_second": 0.402,
+      "step": 84
+    },
+    {
+      "epoch": 1.0714285714285714,
+      "grad_norm": 2.0945916853208053,
+      "learning_rate": 1.7918787065996015e-06,
+      "loss": 0.5453,
+      "step": 90
+    },
+    {
+      "epoch": 1.1904761904761905,
+      "grad_norm": 1.9610907060341172,
+      "learning_rate": 1.7102021003248955e-06,
+      "loss": 0.5353,
+      "step": 100
+    },
+    {
+      "epoch": 1.3095238095238095,
+      "grad_norm": 1.7318461716843294,
+      "learning_rate": 1.6186253687848507e-06,
+      "loss": 0.5312,
+      "step": 110
+    },
+    {
+      "epoch": 1.4285714285714286,
+      "grad_norm": 1.5023311888601967,
+      "learning_rate": 1.5191185619053519e-06,
+      "loss": 0.5261,
+      "step": 120
+    },
+    {
+      "epoch": 1.5476190476190477,
+      "grad_norm": 2.103909947624852,
+      "learning_rate": 1.4138223258333096e-06,
+      "loss": 0.5196,
+      "step": 130
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "grad_norm": 1.4806727889130753,
+      "learning_rate": 1.3050018521581279e-06,
+      "loss": 0.5163,
+      "step": 140
+    },
+    {
+      "epoch": 1.7857142857142856,
+      "grad_norm": 1.7118523951460423,
+      "learning_rate": 1.1949981478418721e-06,
+      "loss": 0.5127,
+      "step": 150
+    },
+    {
+      "epoch": 1.9047619047619047,
+      "grad_norm": 2.1882237413509573,
+      "learning_rate": 1.0861776741666901e-06,
+      "loss": 0.508,
+      "step": 160
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.06450411677360535,
+      "eval_runtime": 89.5353,
+      "eval_samples_per_second": 202.188,
+      "eval_steps_per_second": 0.402,
+      "step": 168
+    },
+    {
+      "epoch": 2.0238095238095237,
+      "grad_norm": 1.7432600091420465,
+      "learning_rate": 9.80881438094648e-07,
+      "loss": 0.5039,
+      "step": 170
+    },
+    {
+      "epoch": 2.142857142857143,
+      "grad_norm": 1.3437595427602773,
+      "learning_rate": 8.813746312151494e-07,
+      "loss": 0.4936,
+      "step": 180
+    },
+    {
+      "epoch": 2.261904761904762,
+      "grad_norm": 1.3903164766916085,
+      "learning_rate": 7.897978996751046e-07,
+      "loss": 0.4913,
+      "step": 190
+    },
+    {
+      "epoch": 2.380952380952381,
+      "grad_norm": 1.4779105389820446,
+      "learning_rate": 7.081212934003984e-07,
+      "loss": 0.487,
+      "step": 200
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 0.8301326396528811,
+      "learning_rate": 6.381018852787821e-07,
+      "loss": 0.4852,
+      "step": 210
+    },
+    {
+      "epoch": 2.619047619047619,
+      "grad_norm": 0.710602110495643,
+      "learning_rate": 5.812459720240681e-07,
+      "loss": 0.4866,
+      "step": 220
+    },
+    {
+      "epoch": 2.738095238095238,
+      "grad_norm": 0.6168443563524292,
+      "learning_rate": 5.387766698777935e-07,
+      "loss": 0.4847,
+      "step": 230
+    },
+    {
+      "epoch": 2.857142857142857,
+      "grad_norm": 0.5597413961712824,
+      "learning_rate": 5.116076022477671e-07,
+      "loss": 0.483,
+      "step": 240
+    },
+    {
+      "epoch": 2.9761904761904763,
+      "grad_norm": 0.5023687044086946,
+      "learning_rate": 5.003232453297512e-07,
+      "loss": 0.48,
+      "step": 250
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 0.06279128044843674,
+      "eval_runtime": 88.7475,
+      "eval_samples_per_second": 203.983,
+      "eval_steps_per_second": 0.406,
+      "step": 252
+    },
+    {
+      "epoch": 3.0,
+      "step": 252,
+      "total_flos": 3375200049561600.0,
+      "train_loss": 0.5529132849640317,
+      "train_runtime": 15188.2312,
+      "train_samples_per_second": 67.938,
+      "train_steps_per_second": 0.017
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 252,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3375200049561600.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}