End of training

README.md

@@ -4,6 +4,7 @@ license: llama3.1
 base_model: meta-llama/Meta-Llama-3.1-8B
 tags:
 - llama-factory
+- full
 - generated_from_trainer
 model-index:
 - name: seed_math_deepmind
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # seed_math_deepmind
 
-This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on an unknown dataset.
+This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on the mlfoundations-dev/seed_math_deepmind dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.1540
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 2.963963963963964,
+    "eval_loss": 0.15399795770645142,
+    "eval_runtime": 30.1298,
+    "eval_samples_per_second": 24.793,
+    "eval_steps_per_second": 0.398,
+    "total_flos": 135468637224960.0,
+    "train_loss": 0.1901383955537537,
+    "train_runtime": 5069.1426,
+    "train_samples_per_second": 8.39,
+    "train_steps_per_second": 0.016
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 2.963963963963964,
+    "eval_loss": 0.15399795770645142,
+    "eval_runtime": 30.1298,
+    "eval_samples_per_second": 24.793,
+    "eval_steps_per_second": 0.398
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.963963963963964,
+    "total_flos": 135468637224960.0,
+    "train_loss": 0.1901383955537537,
+    "train_runtime": 5069.1426,
+    "train_samples_per_second": 8.39,
+    "train_steps_per_second": 0.016
+}

trainer_state.json

@@ -0,0 +1,122 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.963963963963964,
+  "eval_steps": 500,
+  "global_step": 81,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.36036036036036034,
+      "grad_norm": 5.788876433035059,
+      "learning_rate": 5e-06,
+      "loss": 0.3026,
+      "step": 10
+    },
+    {
+      "epoch": 0.7207207207207207,
+      "grad_norm": 2.9871413577007604,
+      "learning_rate": 5e-06,
+      "loss": 0.2081,
+      "step": 20
+    },
+    {
+      "epoch": 0.972972972972973,
+      "eval_loss": 0.18130172789096832,
+      "eval_runtime": 30.3187,
+      "eval_samples_per_second": 24.638,
+      "eval_steps_per_second": 0.396,
+      "step": 27
+    },
+    {
+      "epoch": 1.1036036036036037,
+      "grad_norm": 0.7150273386303733,
+      "learning_rate": 5e-06,
+      "loss": 0.202,
+      "step": 30
+    },
+    {
+      "epoch": 1.4639639639639639,
+      "grad_norm": 0.765396666007098,
+      "learning_rate": 5e-06,
+      "loss": 0.1717,
+      "step": 40
+    },
+    {
+      "epoch": 1.8243243243243243,
+      "grad_norm": 0.43140896524380545,
+      "learning_rate": 5e-06,
+      "loss": 0.1642,
+      "step": 50
+    },
+    {
+      "epoch": 1.9684684684684686,
+      "eval_loss": 0.16076427698135376,
+      "eval_runtime": 30.2833,
+      "eval_samples_per_second": 24.667,
+      "eval_steps_per_second": 0.396,
+      "step": 54
+    },
+    {
+      "epoch": 2.2072072072072073,
+      "grad_norm": 0.44793469318434836,
+      "learning_rate": 5e-06,
+      "loss": 0.172,
+      "step": 60
+    },
+    {
+      "epoch": 2.5675675675675675,
+      "grad_norm": 0.3516467006088966,
+      "learning_rate": 5e-06,
+      "loss": 0.1532,
+      "step": 70
+    },
+    {
+      "epoch": 2.9279279279279278,
+      "grad_norm": 0.49731596311614684,
+      "learning_rate": 5e-06,
+      "loss": 0.1506,
+      "step": 80
+    },
+    {
+      "epoch": 2.963963963963964,
+      "eval_loss": 0.15399795770645142,
+      "eval_runtime": 29.6439,
+      "eval_samples_per_second": 25.199,
+      "eval_steps_per_second": 0.405,
+      "step": 81
+    },
+    {
+      "epoch": 2.963963963963964,
+      "step": 81,
+      "total_flos": 135468637224960.0,
+      "train_loss": 0.1901383955537537,
+      "train_runtime": 5069.1426,
+      "train_samples_per_second": 8.39,
+      "train_steps_per_second": 0.016
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 81,
+  "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": 135468637224960.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}