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
@@ -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
 
-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: 0.0636
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 3.0,
+    "eval_loss": 0.06357227265834808,
+    "eval_runtime": 88.9246,
+    "eval_samples_per_second": 203.577,
+    "eval_steps_per_second": 0.405,
+    "total_flos": 3375200049561600.0,
+    "train_loss": 0.5500972129049755,
+    "train_runtime": 15199.4807,
+    "train_samples_per_second": 67.888,
+    "train_steps_per_second": 0.017
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 3.0,
+    "eval_loss": 0.06357227265834808,
+    "eval_runtime": 88.9246,
+    "eval_samples_per_second": 203.577,
+    "eval_steps_per_second": 0.405
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "total_flos": 3375200049561600.0,
+    "train_loss": 0.5500972129049755,
+    "train_runtime": 15199.4807,
+    "train_samples_per_second": 67.888,
+    "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": 8.739304513169607,
+      "learning_rate": 5.263157894736842e-07,
+      "loss": 0.8721,
+      "step": 10
+    },
+    {
+      "epoch": 0.23809523809523808,
+      "grad_norm": 3.049775296771708,
+      "learning_rate": 1.0526315789473683e-06,
+      "loss": 0.7667,
+      "step": 20
+    },
+    {
+      "epoch": 0.35714285714285715,
+      "grad_norm": 1.4030875344298006,
+      "learning_rate": 1.5789473684210526e-06,
+      "loss": 0.6794,
+      "step": 30
+    },
+    {
+      "epoch": 0.47619047619047616,
+      "grad_norm": 1.3785968299043319,
+      "learning_rate": 1.9995690062269984e-06,
+      "loss": 0.6273,
+      "step": 40
+    },
+    {
+      "epoch": 0.5952380952380952,
+      "grad_norm": 2.874856490919847,
+      "learning_rate": 1.9845231970029773e-06,
+      "loss": 0.5962,
+      "step": 50
+    },
+    {
+      "epoch": 0.7142857142857143,
+      "grad_norm": 1.7727339118671337,
+      "learning_rate": 1.9482977734962752e-06,
+      "loss": 0.5778,
+      "step": 60
+    },
+    {
+      "epoch": 0.8333333333333334,
+      "grad_norm": 1.8345865399680614,
+      "learning_rate": 1.8916720373012423e-06,
+      "loss": 0.5651,
+      "step": 70
+    },
+    {
+      "epoch": 0.9523809523809523,
+      "grad_norm": 1.230639934977736,
+      "learning_rate": 1.8158641529616236e-06,
+      "loss": 0.5532,
+      "step": 80
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.06842269003391266,
+      "eval_runtime": 89.8783,
+      "eval_samples_per_second": 201.417,
+      "eval_steps_per_second": 0.401,
+      "step": 84
+    },
+    {
+      "epoch": 1.0714285714285714,
+      "grad_norm": 1.6345511825327514,
+      "learning_rate": 1.7225049421328022e-06,
+      "loss": 0.5392,
+      "step": 90
+    },
+    {
+      "epoch": 1.1904761904761905,
+      "grad_norm": 1.2835677157215188,
+      "learning_rate": 1.613602800433194e-06,
+      "loss": 0.5279,
+      "step": 100
+    },
+    {
+      "epoch": 1.3095238095238095,
+      "grad_norm": 1.6803587292841158,
+      "learning_rate": 1.4915004917131344e-06,
+      "loss": 0.5234,
+      "step": 110
+    },
+    {
+      "epoch": 1.4285714285714286,
+      "grad_norm": 1.2882453085170966,
+      "learning_rate": 1.3588247492071359e-06,
+      "loss": 0.518,
+      "step": 120
+    },
+    {
+      "epoch": 1.5476190476190477,
+      "grad_norm": 2.32875469479684,
+      "learning_rate": 1.2184297677777462e-06,
+      "loss": 0.5122,
+      "step": 130
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "grad_norm": 1.7960563320982275,
+      "learning_rate": 1.073335802877504e-06,
+      "loss": 0.5096,
+      "step": 140
+    },
+    {
+      "epoch": 1.7857142857142856,
+      "grad_norm": 1.6933464680489778,
+      "learning_rate": 9.266641971224962e-07,
+      "loss": 0.5068,
+      "step": 150
+    },
+    {
+      "epoch": 1.9047619047619047,
+      "grad_norm": 0.9423206924941796,
+      "learning_rate": 7.815702322222537e-07,
+      "loss": 0.5032,
+      "step": 160
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.06419934332370758,
+      "eval_runtime": 89.5881,
+      "eval_samples_per_second": 202.069,
+      "eval_steps_per_second": 0.402,
+      "step": 168
+    },
+    {
+      "epoch": 2.0238095238095237,
+      "grad_norm": 1.0705067821333696,
+      "learning_rate": 6.411752507928641e-07,
+      "loss": 0.4998,
+      "step": 170
+    },
+    {
+      "epoch": 2.142857142857143,
+      "grad_norm": 0.8500177067973432,
+      "learning_rate": 5.084995082868657e-07,
+      "loss": 0.4907,
+      "step": 180
+    },
+    {
+      "epoch": 2.261904761904762,
+      "grad_norm": 0.7596030935231589,
+      "learning_rate": 3.8639719956680615e-07,
+      "loss": 0.4892,
+      "step": 190
+    },
+    {
+      "epoch": 2.380952380952381,
+      "grad_norm": 0.5822861939530503,
+      "learning_rate": 2.7749505786719795e-07,
+      "loss": 0.4856,
+      "step": 200
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 0.5090518617223921,
+      "learning_rate": 1.8413584703837615e-07,
+      "loss": 0.4846,
+      "step": 210
+    },
+    {
+      "epoch": 2.619047619047619,
+      "grad_norm": 0.4618727718359669,
+      "learning_rate": 1.0832796269875755e-07,
+      "loss": 0.4871,
+      "step": 220
+    },
+    {
+      "epoch": 2.738095238095238,
+      "grad_norm": 0.45695002516772665,
+      "learning_rate": 5.170222650372469e-08,
+      "loss": 0.4861,
+      "step": 230
+    },
+    {
+      "epoch": 2.857142857142857,
+      "grad_norm": 0.4368551467326881,
+      "learning_rate": 1.547680299702281e-08,
+      "loss": 0.4855,
+      "step": 240
+    },
+    {
+      "epoch": 2.9761904761904763,
+      "grad_norm": 0.42274290113964946,
+      "learning_rate": 4.309937730015978e-10,
+      "loss": 0.4835,
+      "step": 250
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 0.06357227265834808,
+      "eval_runtime": 88.8886,
+      "eval_samples_per_second": 203.659,
+      "eval_steps_per_second": 0.405,
+      "step": 252
+    },
+    {
+      "epoch": 3.0,
+      "step": 252,
+      "total_flos": 3375200049561600.0,
+      "train_loss": 0.5500972129049755,
+      "train_runtime": 15199.4807,
+      "train_samples_per_second": 67.888,
+      "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
+}