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: oh_v1-2_only_metamath
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # oh_v1-2_only_metamath
 
-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/oh_v1-2_only_metamath dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.4539
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 2.943089430894309,
+    "eval_loss": 0.4539353549480438,
+    "eval_runtime": 11.538,
+    "eval_samples_per_second": 35.708,
+    "eval_steps_per_second": 0.607,
+    "total_flos": 75167296389120.0,
+    "train_loss": 0.5267699400583903,
+    "train_runtime": 1721.4829,
+    "train_samples_per_second": 13.61,
+    "train_steps_per_second": 0.026
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 2.943089430894309,
+    "eval_loss": 0.4539353549480438,
+    "eval_runtime": 11.538,
+    "eval_samples_per_second": 35.708,
+    "eval_steps_per_second": 0.607
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.943089430894309,
+    "total_flos": 75167296389120.0,
+    "train_loss": 0.5267699400583903,
+    "train_runtime": 1721.4829,
+    "train_samples_per_second": 13.61,
+    "train_steps_per_second": 0.026
+}

trainer_state.json

@@ -0,0 +1,94 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.943089430894309,
+  "eval_steps": 500,
+  "global_step": 45,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.6504065040650406,
+      "grad_norm": 1.9171512805599842,
+      "learning_rate": 5e-06,
+      "loss": 0.6422,
+      "step": 10
+    },
+    {
+      "epoch": 0.975609756097561,
+      "eval_loss": 0.5131142735481262,
+      "eval_runtime": 12.7207,
+      "eval_samples_per_second": 32.388,
+      "eval_steps_per_second": 0.55,
+      "step": 15
+    },
+    {
+      "epoch": 1.3089430894308944,
+      "grad_norm": 2.5619927335119006,
+      "learning_rate": 5e-06,
+      "loss": 0.5522,
+      "step": 20
+    },
+    {
+      "epoch": 1.959349593495935,
+      "grad_norm": 0.8478826144949352,
+      "learning_rate": 5e-06,
+      "loss": 0.4763,
+      "step": 30
+    },
+    {
+      "epoch": 1.959349593495935,
+      "eval_loss": 0.4720878303050995,
+      "eval_runtime": 12.2839,
+      "eval_samples_per_second": 33.54,
+      "eval_steps_per_second": 0.57,
+      "step": 30
+    },
+    {
+      "epoch": 2.617886178861789,
+      "grad_norm": 0.740089639808094,
+      "learning_rate": 5e-06,
+      "loss": 0.4862,
+      "step": 40
+    },
+    {
+      "epoch": 2.943089430894309,
+      "eval_loss": 0.4539353549480438,
+      "eval_runtime": 10.807,
+      "eval_samples_per_second": 38.123,
+      "eval_steps_per_second": 0.648,
+      "step": 45
+    },
+    {
+      "epoch": 2.943089430894309,
+      "step": 45,
+      "total_flos": 75167296389120.0,
+      "train_loss": 0.5267699400583903,
+      "train_runtime": 1721.4829,
+      "train_samples_per_second": 13.61,
+      "train_steps_per_second": 0.026
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 45,
+  "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": 75167296389120.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}