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-warmup0.05-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-warmup0.05-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: 0.0622
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 3.0,
+    "eval_loss": 0.062153831124305725,
+    "eval_runtime": 88.9521,
+    "eval_samples_per_second": 203.514,
+    "eval_steps_per_second": 0.405,
+    "total_flos": 3375200049561600.0,
+    "train_loss": 0.5371241434699013,
+    "train_runtime": 15001.3184,
+    "train_samples_per_second": 68.785,
+    "train_steps_per_second": 0.017
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 3.0,
+    "eval_loss": 0.062153831124305725,
+    "eval_runtime": 88.9521,
+    "eval_samples_per_second": 203.514,
+    "eval_steps_per_second": 0.405
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "total_flos": 3375200049561600.0,
+    "train_loss": 0.5371241434699013,
+    "train_runtime": 15001.3184,
+    "train_samples_per_second": 68.785,
+    "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": 6.873670258140521,
+      "learning_rate": 1.5384615384615385e-06,
+      "loss": 0.8434,
+      "step": 10
+    },
+    {
+      "epoch": 0.23809523809523808,
+      "grad_norm": 1.799131772835958,
+      "learning_rate": 1.9968273303105645e-06,
+      "loss": 0.7049,
+      "step": 20
+    },
+    {
+      "epoch": 0.35714285714285715,
+      "grad_norm": 1.7057448542352618,
+      "learning_rate": 1.981352307157077e-06,
+      "loss": 0.6308,
+      "step": 30
+    },
+    {
+      "epoch": 0.47619047619047616,
+      "grad_norm": 3.2000151180883942,
+      "learning_rate": 1.9532588586635555e-06,
+      "loss": 0.5987,
+      "step": 40
+    },
+    {
+      "epoch": 0.5952380952380952,
+      "grad_norm": 1.5365760572953298,
+      "learning_rate": 1.9130316966151297e-06,
+      "loss": 0.5818,
+      "step": 50
+    },
+    {
+      "epoch": 0.7142857142857143,
+      "grad_norm": 2.3881726368494456,
+      "learning_rate": 1.8613648824593137e-06,
+      "loss": 0.5689,
+      "step": 60
+    },
+    {
+      "epoch": 0.8333333333333334,
+      "grad_norm": 1.3104030138998584,
+      "learning_rate": 1.7991498522804724e-06,
+      "loss": 0.5591,
+      "step": 70
+    },
+    {
+      "epoch": 0.9523809523809523,
+      "grad_norm": 1.070092075314844,
+      "learning_rate": 1.7274600363606417e-06,
+      "loss": 0.5488,
+      "step": 80
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.06786686182022095,
+      "eval_runtime": 89.6525,
+      "eval_samples_per_second": 201.924,
+      "eval_steps_per_second": 0.402,
+      "step": 84
+    },
+    {
+      "epoch": 1.0714285714285714,
+      "grad_norm": 1.598258674771628,
+      "learning_rate": 1.6475323386939333e-06,
+      "loss": 0.5342,
+      "step": 90
+    },
+    {
+      "epoch": 1.1904761904761905,
+      "grad_norm": 1.6651178950942467,
+      "learning_rate": 1.560745795998681e-06,
+      "loss": 0.5232,
+      "step": 100
+    },
+    {
+      "epoch": 1.3095238095238095,
+      "grad_norm": 2.0206003862608646,
+      "learning_rate": 1.4685977844351884e-06,
+      "loss": 0.5194,
+      "step": 110
+    },
+    {
+      "epoch": 1.4285714285714286,
+      "grad_norm": 1.5131024957485417,
+      "learning_rate": 1.3726781845477123e-06,
+      "loss": 0.5147,
+      "step": 120
+    },
+    {
+      "epoch": 1.5476190476190477,
+      "grad_norm": 1.0648784167624672,
+      "learning_rate": 1.274641950177214e-06,
+      "loss": 0.5089,
+      "step": 130
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "grad_norm": 1.0063001149457829,
+      "learning_rate": 1.176180554628579e-06,
+      "loss": 0.5062,
+      "step": 140
+    },
+    {
+      "epoch": 1.7857142857142856,
+      "grad_norm": 0.614837213575453,
+      "learning_rate": 1.0789928067473494e-06,
+      "loss": 0.5027,
+      "step": 150
+    },
+    {
+      "epoch": 1.9047619047619047,
+      "grad_norm": 0.6264817095760062,
+      "learning_rate": 9.847555404323228e-07,
+      "loss": 0.4983,
+      "step": 160
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.06357834488153458,
+      "eval_runtime": 89.3936,
+      "eval_samples_per_second": 202.509,
+      "eval_steps_per_second": 0.403,
+      "step": 168
+    },
+    {
+      "epoch": 2.0238095238095237,
+      "grad_norm": 0.7595543213075306,
+      "learning_rate": 8.95094683294048e-07,
+      "loss": 0.4937,
+      "step": 170
+    },
+    {
+      "epoch": 2.142857142857143,
+      "grad_norm": 0.8075613915746389,
+      "learning_rate": 8.115572036276285e-07,
+      "loss": 0.4823,
+      "step": 180
+    },
+    {
+      "epoch": 2.261904761904762,
+      "grad_norm": 0.7283677919022348,
+      "learning_rate": 7.355844197142121e-07,
+      "loss": 0.4802,
+      "step": 190
+    },
+    {
+      "epoch": 2.380952380952381,
+      "grad_norm": 0.5430468412018504,
+      "learning_rate": 6.684871319605308e-07,
+      "loss": 0.4761,
+      "step": 200
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 0.5606001429431275,
+      "learning_rate": 6.114230069354231e-07,
+      "loss": 0.4746,
+      "step": 210
+    },
+    {
+      "epoch": 2.619047619047619,
+      "grad_norm": 0.5550155038305291,
+      "learning_rate": 5.653766035090493e-07,
+      "loss": 0.4764,
+      "step": 220
+    },
+    {
+      "epoch": 2.738095238095238,
+      "grad_norm": 0.579557506598742,
+      "learning_rate": 5.311423857148533e-07,
+      "loss": 0.4747,
+      "step": 230
+    },
+    {
+      "epoch": 2.857142857142857,
+      "grad_norm": 0.5367862820074156,
+      "learning_rate": 5.093110154227483e-07,
+      "loss": 0.4734,
+      "step": 240
+    },
+    {
+      "epoch": 2.9761904761904763,
+      "grad_norm": 0.5135404834401238,
+      "learning_rate": 5.002591613236042e-07,
+      "loss": 0.4707,
+      "step": 250
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 0.062153831124305725,
+      "eval_runtime": 89.1749,
+      "eval_samples_per_second": 203.006,
+      "eval_steps_per_second": 0.404,
+      "step": 252
+    },
+    {
+      "epoch": 3.0,
+      "step": 252,
+      "total_flos": 3375200049561600.0,
+      "train_loss": 0.5371241434699013,
+      "train_runtime": 15001.3184,
+      "train_samples_per_second": 68.785,
+      "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
+}