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_chinese
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # stackexchange_chinese
 
-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_chinese dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.8938
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 2.9217391304347826,
+    "eval_loss": 0.8937940001487732,
+    "eval_runtime": 17.3865,
+    "eval_samples_per_second": 22.259,
+    "eval_steps_per_second": 0.403,
+    "total_flos": 70142184652800.0,
+    "train_loss": 0.997152254694984,
+    "train_runtime": 2754.5672,
+    "train_samples_per_second": 7.993,
+    "train_steps_per_second": 0.015
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 2.9217391304347826,
+    "eval_loss": 0.8937940001487732,
+    "eval_runtime": 17.3865,
+    "eval_samples_per_second": 22.259,
+    "eval_steps_per_second": 0.403
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.9217391304347826,
+    "total_flos": 70142184652800.0,
+    "train_loss": 0.997152254694984,
+    "train_runtime": 2754.5672,
+    "train_samples_per_second": 7.993,
+    "train_steps_per_second": 0.015
+}

trainer_state.json

@@ -0,0 +1,94 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.9217391304347826,
+  "eval_steps": 500,
+  "global_step": 42,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.6956521739130435,
+      "grad_norm": 8.691084236878861,
+      "learning_rate": 5e-06,
+      "loss": 1.1365,
+      "step": 10
+    },
+    {
+      "epoch": 0.9739130434782609,
+      "eval_loss": 1.0025559663772583,
+      "eval_runtime": 17.7051,
+      "eval_samples_per_second": 21.858,
+      "eval_steps_per_second": 0.395,
+      "step": 14
+    },
+    {
+      "epoch": 1.391304347826087,
+      "grad_norm": 6.0498570349381575,
+      "learning_rate": 5e-06,
+      "loss": 1.0486,
+      "step": 20
+    },
+    {
+      "epoch": 1.9478260869565216,
+      "eval_loss": 0.9321419596672058,
+      "eval_runtime": 17.6501,
+      "eval_samples_per_second": 21.926,
+      "eval_steps_per_second": 0.397,
+      "step": 28
+    },
+    {
+      "epoch": 2.0869565217391304,
+      "grad_norm": 1.9902042350733853,
+      "learning_rate": 5e-06,
+      "loss": 0.9777,
+      "step": 30
+    },
+    {
+      "epoch": 2.782608695652174,
+      "grad_norm": 1.5660035860374173,
+      "learning_rate": 5e-06,
+      "loss": 0.8562,
+      "step": 40
+    },
+    {
+      "epoch": 2.9217391304347826,
+      "eval_loss": 0.8937940001487732,
+      "eval_runtime": 16.9743,
+      "eval_samples_per_second": 22.799,
+      "eval_steps_per_second": 0.412,
+      "step": 42
+    },
+    {
+      "epoch": 2.9217391304347826,
+      "step": 42,
+      "total_flos": 70142184652800.0,
+      "train_loss": 0.997152254694984,
+      "train_runtime": 2754.5672,
+      "train_samples_per_second": 7.993,
+      "train_steps_per_second": 0.015
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 42,
+  "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": 70142184652800.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}