End of training

README.md

@@ -4,6 +4,7 @@ library_name: peft
 license: llama3.1
 tags:
 - llama-factory
+- lora
 - generated_from_trainer
 model-index:
 - name: Llama-3.1-8B-Instruct-SFT-100
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # Llama-3.1-8B-Instruct-SFT-100
 
-This model is a fine-tuned version of [meta-llama/Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct) on an unknown dataset.
+This model is a fine-tuned version of [meta-llama/Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct) on the bct_non_cot_sft_100 dataset.
 It achieves the following results on the evaluation set:
 - Loss: 1.1914
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 8.88888888888889,
+    "eval_loss": 1.191369891166687,
+    "eval_runtime": 0.2544,
+    "eval_samples_per_second": 39.311,
+    "eval_steps_per_second": 19.656,
+    "total_flos": 4594288692953088.0,
+    "train_loss": 1.2649920654296876,
+    "train_runtime": 69.6409,
+    "train_samples_per_second": 12.923,
+    "train_steps_per_second": 0.718
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 8.88888888888889,
+    "eval_loss": 1.191369891166687,
+    "eval_runtime": 0.2544,
+    "eval_samples_per_second": 39.311,
+    "eval_steps_per_second": 19.656
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 8.88888888888889,
+    "total_flos": 4594288692953088.0,
+    "train_loss": 1.2649920654296876,
+    "train_runtime": 69.6409,
+    "train_samples_per_second": 12.923,
+    "train_steps_per_second": 0.718
+}

trainer_state.json

@@ -0,0 +1,85 @@
+{
+  "best_metric": 1.191369891166687,
+  "best_model_checkpoint": "saves/Llama-3.1-8B-Instruct/lora/sft-100/checkpoint-50",
+  "epoch": 8.88888888888889,
+  "eval_steps": 50,
+  "global_step": 50,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.7777777777777777,
+      "grad_norm": 6.377140522003174,
+      "learning_rate": 4.849231551964771e-06,
+      "loss": 1.6881,
+      "step": 10
+    },
+    {
+      "epoch": 3.5555555555555554,
+      "grad_norm": 5.378269195556641,
+      "learning_rate": 3.7500000000000005e-06,
+      "loss": 1.4551,
+      "step": 20
+    },
+    {
+      "epoch": 5.333333333333333,
+      "grad_norm": 3.855208158493042,
+      "learning_rate": 2.0658795558326745e-06,
+      "loss": 1.1462,
+      "step": 30
+    },
+    {
+      "epoch": 7.111111111111111,
+      "grad_norm": 3.9790968894958496,
+      "learning_rate": 5.848888922025553e-07,
+      "loss": 1.0317,
+      "step": 40
+    },
+    {
+      "epoch": 8.88888888888889,
+      "grad_norm": 3.3717496395111084,
+      "learning_rate": 6.089874350439507e-09,
+      "loss": 1.0038,
+      "step": 50
+    },
+    {
+      "epoch": 8.88888888888889,
+      "eval_loss": 1.191369891166687,
+      "eval_runtime": 0.2516,
+      "eval_samples_per_second": 39.753,
+      "eval_steps_per_second": 19.877,
+      "step": 50
+    },
+    {
+      "epoch": 8.88888888888889,
+      "step": 50,
+      "total_flos": 4594288692953088.0,
+      "train_loss": 1.2649920654296876,
+      "train_runtime": 69.6409,
+      "train_samples_per_second": 12.923,
+      "train_steps_per_second": 0.718
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 50,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 50,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4594288692953088.0,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}