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: llama3-1_8b_4o_annotated_aops
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # llama3-1_8b_4o_annotated_aops
 
-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/4o_annotated_aops dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.7526
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 3.0,
+    "eval_loss": 0.7526352405548096,
+    "eval_runtime": 1.839,
+    "eval_samples_per_second": 238.72,
+    "eval_steps_per_second": 1.088,
+    "total_flos": 85426899517440.0,
+    "train_loss": 0.7682641569305869,
+    "train_runtime": 638.061,
+    "train_samples_per_second": 39.217,
+    "train_steps_per_second": 0.08
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 3.0,
+    "eval_loss": 0.7526352405548096,
+    "eval_runtime": 1.839,
+    "eval_samples_per_second": 238.72,
+    "eval_steps_per_second": 1.088
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "total_flos": 85426899517440.0,
+    "train_loss": 0.7682641569305869,
+    "train_runtime": 638.061,
+    "train_samples_per_second": 39.217,
+    "train_steps_per_second": 0.08
+}

trainer_state.json

@@ -0,0 +1,101 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.0,
+  "eval_steps": 500,
+  "global_step": 51,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.5882352941176471,
+      "grad_norm": 3.66774997978251,
+      "learning_rate": 5e-06,
+      "loss": 0.8804,
+      "step": 10
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.7954549193382263,
+      "eval_runtime": 1.8752,
+      "eval_samples_per_second": 234.109,
+      "eval_steps_per_second": 1.067,
+      "step": 17
+    },
+    {
+      "epoch": 1.1764705882352942,
+      "grad_norm": 2.6669575605278664,
+      "learning_rate": 5e-06,
+      "loss": 0.7874,
+      "step": 20
+    },
+    {
+      "epoch": 1.7647058823529411,
+      "grad_norm": 1.184947489520886,
+      "learning_rate": 5e-06,
+      "loss": 0.7508,
+      "step": 30
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.7634365558624268,
+      "eval_runtime": 1.8503,
+      "eval_samples_per_second": 237.259,
+      "eval_steps_per_second": 1.081,
+      "step": 34
+    },
+    {
+      "epoch": 2.3529411764705883,
+      "grad_norm": 0.48400852002984446,
+      "learning_rate": 5e-06,
+      "loss": 0.7209,
+      "step": 40
+    },
+    {
+      "epoch": 2.9411764705882355,
+      "grad_norm": 0.6808590644241648,
+      "learning_rate": 5e-06,
+      "loss": 0.7022,
+      "step": 50
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 0.7526352405548096,
+      "eval_runtime": 4.0053,
+      "eval_samples_per_second": 109.604,
+      "eval_steps_per_second": 0.499,
+      "step": 51
+    },
+    {
+      "epoch": 3.0,
+      "step": 51,
+      "total_flos": 85426899517440.0,
+      "train_loss": 0.7682641569305869,
+      "train_runtime": 638.061,
+      "train_samples_per_second": 39.217,
+      "train_steps_per_second": 0.08
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 51,
+  "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": 85426899517440.0,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}