End of training

README.md

@@ -4,6 +4,7 @@ library_name: peft
 license: gemma
 tags:
 - llama-factory
+- lora
 - generated_from_trainer
 model-index:
 - name: Gemma-2-9B-It-SFT-Half
@@ -15,7 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # Gemma-2-9B-It-SFT-Half
 
-This model is a fine-tuned version of [google/gemma-2-9b-it](https://huggingface.co/google/gemma-2-9b-it) on an unknown dataset.
+This model is a fine-tuned version of [google/gemma-2-9b-it](https://huggingface.co/google/gemma-2-9b-it) on the bct_non_cot_sft_500 dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.0125
 
 ## Model description
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 2.986666666666667,
+    "eval_loss": 1.0124613046646118,
+    "eval_runtime": 3.0254,
+    "eval_samples_per_second": 16.527,
+    "eval_steps_per_second": 8.263,
+    "total_flos": 8925507086598144.0,
+    "train_loss": 1.605351448059082,
+    "train_runtime": 305.2742,
+    "train_samples_per_second": 4.422,
+    "train_steps_per_second": 0.275
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 2.986666666666667,
+    "eval_loss": 1.0124613046646118,
+    "eval_runtime": 3.0254,
+    "eval_samples_per_second": 16.527,
+    "eval_steps_per_second": 8.263
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.986666666666667,
+    "total_flos": 8925507086598144.0,
+    "train_loss": 1.605351448059082,
+    "train_runtime": 305.2742,
+    "train_samples_per_second": 4.422,
+    "train_steps_per_second": 0.275
+}

trainer_state.json

@@ -0,0 +1,98 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.986666666666667,
+  "eval_steps": 500,
+  "global_step": 84,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.35555555555555557,
+      "grad_norm": 7.270255088806152,
+      "learning_rate": 4.997807075247147e-06,
+      "loss": 2.8249,
+      "step": 10
+    },
+    {
+      "epoch": 0.7111111111111111,
+      "grad_norm": 7.416280746459961,
+      "learning_rate": 4.7392794005985324e-06,
+      "loss": 2.2568,
+      "step": 20
+    },
+    {
+      "epoch": 1.0666666666666667,
+      "grad_norm": 5.478653430938721,
+      "learning_rate": 4.093559974371725e-06,
+      "loss": 1.9466,
+      "step": 30
+    },
+    {
+      "epoch": 1.4222222222222223,
+      "grad_norm": 5.283214569091797,
+      "learning_rate": 3.1722995515381644e-06,
+      "loss": 1.5913,
+      "step": 40
+    },
+    {
+      "epoch": 1.7777777777777777,
+      "grad_norm": 4.807966232299805,
+      "learning_rate": 2.134792428593971e-06,
+      "loss": 1.2558,
+      "step": 50
+    },
+    {
+      "epoch": 2.1333333333333333,
+      "grad_norm": 4.748135089874268,
+      "learning_rate": 1.160433012552508e-06,
+      "loss": 1.1178,
+      "step": 60
+    },
+    {
+      "epoch": 2.488888888888889,
+      "grad_norm": 4.198294162750244,
+      "learning_rate": 4.1769689822475147e-07,
+      "loss": 1.0357,
+      "step": 70
+    },
+    {
+      "epoch": 2.8444444444444446,
+      "grad_norm": 5.134349822998047,
+      "learning_rate": 3.5009907323737826e-08,
+      "loss": 1.008,
+      "step": 80
+    },
+    {
+      "epoch": 2.986666666666667,
+      "step": 84,
+      "total_flos": 8925507086598144.0,
+      "train_loss": 1.605351448059082,
+      "train_runtime": 305.2742,
+      "train_samples_per_second": 4.422,
+      "train_steps_per_second": 0.275
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 84,
+  "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": 8925507086598144.0,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}