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_lr5e-6_scheduler-cosine-warmup0.05-minlr5e-7
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # hp_ablations_grid_mistral_bsz4096_lr5e-6_scheduler-cosine-warmup0.05-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: 0.0594
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 3.0,
+    "eval_loss": 0.05939801409840584,
+    "eval_runtime": 88.9844,
+    "eval_samples_per_second": 203.44,
+    "eval_steps_per_second": 0.405,
+    "total_flos": 3375200049561600.0,
+    "train_loss": 0.49902858005629647,
+    "train_runtime": 15008.2789,
+    "train_samples_per_second": 68.753,
+    "train_steps_per_second": 0.017
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 3.0,
+    "eval_loss": 0.05939801409840584,
+    "eval_runtime": 88.9844,
+    "eval_samples_per_second": 203.44,
+    "eval_steps_per_second": 0.405
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "total_flos": 3375200049561600.0,
+    "train_loss": 0.49902858005629647,
+    "train_runtime": 15008.2789,
+    "train_samples_per_second": 68.753,
+    "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": 6.199301191341975,
+      "learning_rate": 3.846153846153847e-06,
+      "loss": 0.8108,
+      "step": 10
+    },
+    {
+      "epoch": 0.23809523809523808,
+      "grad_norm": 2.652045367040131,
+      "learning_rate": 4.990481990931694e-06,
+      "loss": 0.6769,
+      "step": 20
+    },
+    {
+      "epoch": 0.35714285714285715,
+      "grad_norm": 2.723155487023949,
+      "learning_rate": 4.944056921471231e-06,
+      "loss": 0.6055,
+      "step": 30
+    },
+    {
+      "epoch": 0.47619047619047616,
+      "grad_norm": 1.9826207941632754,
+      "learning_rate": 4.859776575990668e-06,
+      "loss": 0.5706,
+      "step": 40
+    },
+    {
+      "epoch": 0.5952380952380952,
+      "grad_norm": 3.5654763295831517,
+      "learning_rate": 4.73909508984539e-06,
+      "loss": 0.5522,
+      "step": 50
+    },
+    {
+      "epoch": 0.7142857142857143,
+      "grad_norm": 1.8104618201572298,
+      "learning_rate": 4.5840946473779415e-06,
+      "loss": 0.5376,
+      "step": 60
+    },
+    {
+      "epoch": 0.8333333333333334,
+      "grad_norm": 1.7247740898728032,
+      "learning_rate": 4.397449556841417e-06,
+      "loss": 0.5269,
+      "step": 70
+    },
+    {
+      "epoch": 0.9523809523809523,
+      "grad_norm": 1.7338607879041106,
+      "learning_rate": 4.1823801090819264e-06,
+      "loss": 0.518,
+      "step": 80
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.06437206268310547,
+      "eval_runtime": 89.36,
+      "eval_samples_per_second": 202.585,
+      "eval_steps_per_second": 0.403,
+      "step": 84
+    },
+    {
+      "epoch": 1.0714285714285714,
+      "grad_norm": 2.365498030856518,
+      "learning_rate": 3.9425970160818e-06,
+      "loss": 0.4982,
+      "step": 90
+    },
+    {
+      "epoch": 1.1904761904761905,
+      "grad_norm": 2.9959916872592225,
+      "learning_rate": 3.6822373879960433e-06,
+      "loss": 0.4824,
+      "step": 100
+    },
+    {
+      "epoch": 1.3095238095238095,
+      "grad_norm": 2.4377878704528806,
+      "learning_rate": 3.4057933533055654e-06,
+      "loss": 0.4791,
+      "step": 110
+    },
+    {
+      "epoch": 1.4285714285714286,
+      "grad_norm": 2.2246867957095815,
+      "learning_rate": 3.118034553643137e-06,
+      "loss": 0.4752,
+      "step": 120
+    },
+    {
+      "epoch": 1.5476190476190477,
+      "grad_norm": 1.4481256786493237,
+      "learning_rate": 2.8239258505316425e-06,
+      "loss": 0.4698,
+      "step": 130
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "grad_norm": 1.8193175030418456,
+      "learning_rate": 2.5285416638857373e-06,
+      "loss": 0.4679,
+      "step": 140
+    },
+    {
+      "epoch": 1.7857142857142856,
+      "grad_norm": 2.466154702975751,
+      "learning_rate": 2.236978420242048e-06,
+      "loss": 0.4652,
+      "step": 150
+    },
+    {
+      "epoch": 1.9047619047619047,
+      "grad_norm": 1.147751249178101,
+      "learning_rate": 1.9542666212969685e-06,
+      "loss": 0.4614,
+      "step": 160
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.06041712313890457,
+      "eval_runtime": 89.2936,
+      "eval_samples_per_second": 202.736,
+      "eval_steps_per_second": 0.403,
+      "step": 168
+    },
+    {
+      "epoch": 2.0238095238095237,
+      "grad_norm": 1.4837181107165451,
+      "learning_rate": 1.6852840498821444e-06,
+      "loss": 0.4551,
+      "step": 170
+    },
+    {
+      "epoch": 2.142857142857143,
+      "grad_norm": 1.1499094737034221,
+      "learning_rate": 1.4346716108828857e-06,
+      "loss": 0.4354,
+      "step": 180
+    },
+    {
+      "epoch": 2.261904761904762,
+      "grad_norm": 0.8873968255806381,
+      "learning_rate": 1.2067532591426361e-06,
+      "loss": 0.4335,
+      "step": 190
+    },
+    {
+      "epoch": 2.380952380952381,
+      "grad_norm": 0.6497450675540968,
+      "learning_rate": 1.0054613958815923e-06,
+      "loss": 0.43,
+      "step": 200
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 0.543195722444911,
+      "learning_rate": 8.342690208062692e-07,
+      "loss": 0.4288,
+      "step": 210
+    },
+    {
+      "epoch": 2.619047619047619,
+      "grad_norm": 0.5271957215457769,
+      "learning_rate": 6.961298105271477e-07,
+      "loss": 0.4306,
+      "step": 220
+    },
+    {
+      "epoch": 2.738095238095238,
+      "grad_norm": 0.540487189738211,
+      "learning_rate": 5.9342715714456e-07,
+      "loss": 0.4294,
+      "step": 230
+    },
+    {
+      "epoch": 2.857142857142857,
+      "grad_norm": 0.5473176030765594,
+      "learning_rate": 5.27933046268245e-07,
+      "loss": 0.4283,
+      "step": 240
+    },
+    {
+      "epoch": 2.9761904761904763,
+      "grad_norm": 0.47492816840323937,
+      "learning_rate": 5.007774839708124e-07,
+      "loss": 0.426,
+      "step": 250
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 0.05939801409840584,
+      "eval_runtime": 88.7716,
+      "eval_samples_per_second": 203.928,
+      "eval_steps_per_second": 0.406,
+      "step": 252
+    },
+    {
+      "epoch": 3.0,
+      "step": 252,
+      "total_flos": 3375200049561600.0,
+      "train_loss": 0.49902858005629647,
+      "train_runtime": 15008.2789,
+      "train_samples_per_second": 68.753,
+      "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
+}