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: stackexchange_chess
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # stackexchange_chess
 
-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/stackexchange_chess dataset.
 It achieves the following results on the evaluation set:
 - Loss: 1.0862
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 2.888888888888889,
+    "eval_loss": 1.0861635208129883,
+    "eval_runtime": 19.8419,
+    "eval_samples_per_second": 22.78,
+    "eval_steps_per_second": 0.403,
+    "total_flos": 80192408125440.0,
+    "train_loss": 1.1697100400924683,
+    "train_runtime": 3119.1782,
+    "train_samples_per_second": 8.252,
+    "train_steps_per_second": 0.015
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 2.888888888888889,
+    "eval_loss": 1.0861635208129883,
+    "eval_runtime": 19.8419,
+    "eval_samples_per_second": 22.78,
+    "eval_steps_per_second": 0.403
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.888888888888889,
+    "total_flos": 80192408125440.0,
+    "train_loss": 1.1697100400924683,
+    "train_runtime": 3119.1782,
+    "train_samples_per_second": 8.252,
+    "train_steps_per_second": 0.015
+}

trainer_state.json

@@ -0,0 +1,94 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.888888888888889,
+  "eval_steps": 500,
+  "global_step": 48,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.5925925925925926,
+      "grad_norm": 14.218648414682958,
+      "learning_rate": 5e-06,
+      "loss": 1.3225,
+      "step": 10
+    },
+    {
+      "epoch": 0.9481481481481482,
+      "eval_loss": 1.1893168687820435,
+      "eval_runtime": 20.3378,
+      "eval_samples_per_second": 22.225,
+      "eval_steps_per_second": 0.393,
+      "step": 16
+    },
+    {
+      "epoch": 1.2074074074074075,
+      "grad_norm": 3.825380635531979,
+      "learning_rate": 5e-06,
+      "loss": 1.2402,
+      "step": 20
+    },
+    {
+      "epoch": 1.8,
+      "grad_norm": 2.1171757799420057,
+      "learning_rate": 5e-06,
+      "loss": 1.115,
+      "step": 30
+    },
+    {
+      "epoch": 1.9777777777777779,
+      "eval_loss": 1.117397665977478,
+      "eval_runtime": 20.2101,
+      "eval_samples_per_second": 22.365,
+      "eval_steps_per_second": 0.396,
+      "step": 33
+    },
+    {
+      "epoch": 2.414814814814815,
+      "grad_norm": 1.868144898782511,
+      "learning_rate": 5e-06,
+      "loss": 1.1122,
+      "step": 40
+    },
+    {
+      "epoch": 2.888888888888889,
+      "eval_loss": 1.0861635208129883,
+      "eval_runtime": 19.4458,
+      "eval_samples_per_second": 23.244,
+      "eval_steps_per_second": 0.411,
+      "step": 48
+    },
+    {
+      "epoch": 2.888888888888889,
+      "step": 48,
+      "total_flos": 80192408125440.0,
+      "train_loss": 1.1697100400924683,
+      "train_runtime": 3119.1782,
+      "train_samples_per_second": 8.252,
+      "train_steps_per_second": 0.015
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 48,
+  "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": 80192408125440.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}