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_automathtext
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # seed_math_automathtext
 
-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_automathtext dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.4801
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 3.0,
+    "eval_loss": 0.4801027178764343,
+    "eval_runtime": 42.3155,
+    "eval_samples_per_second": 24.79,
+    "eval_steps_per_second": 0.402,
+    "total_flos": 195769978060800.0,
+    "train_loss": 0.4775498178270128,
+    "train_runtime": 7148.9147,
+    "train_samples_per_second": 8.356,
+    "train_steps_per_second": 0.016
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 3.0,
+    "eval_loss": 0.4801027178764343,
+    "eval_runtime": 42.3155,
+    "eval_samples_per_second": 24.79,
+    "eval_steps_per_second": 0.402
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "total_flos": 195769978060800.0,
+    "train_loss": 0.4775498178270128,
+    "train_runtime": 7148.9147,
+    "train_samples_per_second": 8.356,
+    "train_steps_per_second": 0.016
+}

trainer_state.json

@@ -0,0 +1,143 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.0,
+  "eval_steps": 500,
+  "global_step": 117,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.2564102564102564,
+      "grad_norm": 1.7402502022772977,
+      "learning_rate": 5e-06,
+      "loss": 0.6037,
+      "step": 10
+    },
+    {
+      "epoch": 0.5128205128205128,
+      "grad_norm": 1.5163040490783106,
+      "learning_rate": 5e-06,
+      "loss": 0.5362,
+      "step": 20
+    },
+    {
+      "epoch": 0.7692307692307693,
+      "grad_norm": 0.4907623553031213,
+      "learning_rate": 5e-06,
+      "loss": 0.5112,
+      "step": 30
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.4971420168876648,
+      "eval_runtime": 42.8991,
+      "eval_samples_per_second": 24.453,
+      "eval_steps_per_second": 0.396,
+      "step": 39
+    },
+    {
+      "epoch": 1.0256410256410255,
+      "grad_norm": 0.5535851468329118,
+      "learning_rate": 5e-06,
+      "loss": 0.4974,
+      "step": 40
+    },
+    {
+      "epoch": 1.282051282051282,
+      "grad_norm": 0.4334428968066213,
+      "learning_rate": 5e-06,
+      "loss": 0.4671,
+      "step": 50
+    },
+    {
+      "epoch": 1.5384615384615383,
+      "grad_norm": 0.5688870905744688,
+      "learning_rate": 5e-06,
+      "loss": 0.47,
+      "step": 60
+    },
+    {
+      "epoch": 1.7948717948717947,
+      "grad_norm": 0.560760293107518,
+      "learning_rate": 5e-06,
+      "loss": 0.4657,
+      "step": 70
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.48121070861816406,
+      "eval_runtime": 42.5529,
+      "eval_samples_per_second": 24.652,
+      "eval_steps_per_second": 0.4,
+      "step": 78
+    },
+    {
+      "epoch": 2.051282051282051,
+      "grad_norm": 0.7390965783476714,
+      "learning_rate": 5e-06,
+      "loss": 0.4509,
+      "step": 80
+    },
+    {
+      "epoch": 2.3076923076923075,
+      "grad_norm": 0.6414316859859929,
+      "learning_rate": 5e-06,
+      "loss": 0.4352,
+      "step": 90
+    },
+    {
+      "epoch": 2.564102564102564,
+      "grad_norm": 0.5549388627791971,
+      "learning_rate": 5e-06,
+      "loss": 0.4268,
+      "step": 100
+    },
+    {
+      "epoch": 2.8205128205128203,
+      "grad_norm": 0.5228337642638744,
+      "learning_rate": 5e-06,
+      "loss": 0.4298,
+      "step": 110
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 0.4801027178764343,
+      "eval_runtime": 41.9671,
+      "eval_samples_per_second": 24.996,
+      "eval_steps_per_second": 0.405,
+      "step": 117
+    },
+    {
+      "epoch": 3.0,
+      "step": 117,
+      "total_flos": 195769978060800.0,
+      "train_loss": 0.4775498178270128,
+      "train_runtime": 7148.9147,
+      "train_samples_per_second": 8.356,
+      "train_steps_per_second": 0.016
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 117,
+  "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": 195769978060800.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}