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_math_instruct
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # seed_math_math_instruct
 
-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_math_instruct dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.2846
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 3.0,
+    "eval_loss": 0.2846110165119171,
+    "eval_runtime": 22.5392,
+    "eval_samples_per_second": 23.914,
+    "eval_steps_per_second": 0.399,
+    "total_flos": 100292855070720.0,
+    "train_loss": 0.3573924700419108,
+    "train_runtime": 3853.4509,
+    "train_samples_per_second": 7.96,
+    "train_steps_per_second": 0.016
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 3.0,
+    "eval_loss": 0.2846110165119171,
+    "eval_runtime": 22.5392,
+    "eval_samples_per_second": 23.914,
+    "eval_steps_per_second": 0.399
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "total_flos": 100292855070720.0,
+    "train_loss": 0.3573924700419108,
+    "train_runtime": 3853.4509,
+    "train_samples_per_second": 7.96,
+    "train_steps_per_second": 0.016
+}

trainer_state.json

@@ -0,0 +1,108 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.0,
+  "eval_steps": 500,
+  "global_step": 60,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.5,
+      "grad_norm": 9.754155429079685,
+      "learning_rate": 5e-06,
+      "loss": 0.523,
+      "step": 10
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.857858513134004,
+      "learning_rate": 5e-06,
+      "loss": 0.386,
+      "step": 20
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.34043166041374207,
+      "eval_runtime": 22.8091,
+      "eval_samples_per_second": 23.631,
+      "eval_steps_per_second": 0.395,
+      "step": 20
+    },
+    {
+      "epoch": 1.5,
+      "grad_norm": 0.7001146076460096,
+      "learning_rate": 5e-06,
+      "loss": 0.3382,
+      "step": 30
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.9539021480035809,
+      "learning_rate": 5e-06,
+      "loss": 0.3162,
+      "step": 40
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.3004086911678314,
+      "eval_runtime": 22.9234,
+      "eval_samples_per_second": 23.513,
+      "eval_steps_per_second": 0.393,
+      "step": 40
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 0.8380959453139918,
+      "learning_rate": 5e-06,
+      "loss": 0.2959,
+      "step": 50
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.6071539034449804,
+      "learning_rate": 5e-06,
+      "loss": 0.2852,
+      "step": 60
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 0.2846110165119171,
+      "eval_runtime": 21.938,
+      "eval_samples_per_second": 24.569,
+      "eval_steps_per_second": 0.41,
+      "step": 60
+    },
+    {
+      "epoch": 3.0,
+      "step": 60,
+      "total_flos": 100292855070720.0,
+      "train_loss": 0.3573924700419108,
+      "train_runtime": 3853.4509,
+      "train_samples_per_second": 7.96,
+      "train_steps_per_second": 0.016
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 60,
+  "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": 100292855070720.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}