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_r1_annotated_math
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # llama3-1_8b_r1_annotated_math
 
-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/r1_annotated_math dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.5834
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 3.0,
+    "eval_loss": 0.5833666324615479,
+    "eval_runtime": 1.8372,
+    "eval_samples_per_second": 162.199,
+    "eval_steps_per_second": 1.089,
+    "total_flos": 60301340835840.0,
+    "train_loss": 0.6545261012183295,
+    "train_runtime": 627.5494,
+    "train_samples_per_second": 27.043,
+    "train_steps_per_second": 0.057
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 3.0,
+    "eval_loss": 0.5833666324615479,
+    "eval_runtime": 1.8372,
+    "eval_samples_per_second": 162.199,
+    "eval_steps_per_second": 1.089
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "total_flos": 60301340835840.0,
+    "train_loss": 0.6545261012183295,
+    "train_runtime": 627.5494,
+    "train_samples_per_second": 27.043,
+    "train_steps_per_second": 0.057
+}

trainer_state.json

@@ -0,0 +1,87 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.0,
+  "eval_steps": 500,
+  "global_step": 36,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.8333333333333334,
+      "grad_norm": 1.4976304773590843,
+      "learning_rate": 5e-06,
+      "loss": 0.8098,
+      "step": 10
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.6748126149177551,
+      "eval_runtime": 1.8628,
+      "eval_samples_per_second": 159.978,
+      "eval_steps_per_second": 1.074,
+      "step": 12
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "grad_norm": 0.9879481748413028,
+      "learning_rate": 5e-06,
+      "loss": 0.6489,
+      "step": 20
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.6079719066619873,
+      "eval_runtime": 1.8561,
+      "eval_samples_per_second": 160.554,
+      "eval_steps_per_second": 1.078,
+      "step": 24
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 1.0949275268721474,
+      "learning_rate": 5e-06,
+      "loss": 0.5774,
+      "step": 30
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 0.5833666324615479,
+      "eval_runtime": 1.8239,
+      "eval_samples_per_second": 163.388,
+      "eval_steps_per_second": 1.097,
+      "step": 36
+    },
+    {
+      "epoch": 3.0,
+      "step": 36,
+      "total_flos": 60301340835840.0,
+      "train_loss": 0.6545261012183295,
+      "train_runtime": 627.5494,
+      "train_samples_per_second": 27.043,
+      "train_steps_per_second": 0.057
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 36,
+  "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": 60301340835840.0,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}