End of training

README.md

@@ -3,6 +3,7 @@ license: apache-2.0
 library_name: peft
 tags:
 - llama-factory
+- lora
 - generated_from_trainer
 base_model: mistralai/Mistral-7B-Instruct-v0.2
 model-index:
@@ -15,7 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # Mistral-7B-Instruct-v0.2-ORPO-SFT
 
-This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) on an unknown dataset.
+This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) on the bct_non_cot_sft_1000 dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.0648
 
 ## Model description
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 2.986666666666667,
+    "eval_loss": 0.06476102769374847,
+    "eval_runtime": 3.2886,
+    "eval_samples_per_second": 30.408,
+    "eval_steps_per_second": 15.204,
+    "total_flos": 1.6019812625547264e+16,
+    "train_loss": 0.12199947628236953,
+    "train_runtime": 363.0569,
+    "train_samples_per_second": 7.437,
+    "train_steps_per_second": 0.463
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 2.986666666666667,
+    "eval_loss": 0.06476102769374847,
+    "eval_runtime": 3.2886,
+    "eval_samples_per_second": 30.408,
+    "eval_steps_per_second": 15.204
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.986666666666667,
+    "total_flos": 1.6019812625547264e+16,
+    "train_loss": 0.12199947628236953,
+    "train_runtime": 363.0569,
+    "train_samples_per_second": 7.437,
+    "train_steps_per_second": 0.463
+}

trainer_state.json

@@ -0,0 +1,142 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.986666666666667,
+  "eval_steps": 500,
+  "global_step": 168,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.17777777777777778,
+      "grad_norm": 4.85897159576416,
+      "learning_rate": 4.957230266673969e-06,
+      "loss": 0.556,
+      "step": 10
+    },
+    {
+      "epoch": 0.35555555555555557,
+      "grad_norm": 4.12825870513916,
+      "learning_rate": 4.828686741593921e-06,
+      "loss": 0.2945,
+      "step": 20
+    },
+    {
+      "epoch": 0.5333333333333333,
+      "grad_norm": 2.6030149459838867,
+      "learning_rate": 4.618852307232078e-06,
+      "loss": 0.1525,
+      "step": 30
+    },
+    {
+      "epoch": 0.7111111111111111,
+      "grad_norm": 3.3148317337036133,
+      "learning_rate": 4.335051964269395e-06,
+      "loss": 0.0805,
+      "step": 40
+    },
+    {
+      "epoch": 0.8888888888888888,
+      "grad_norm": 2.82584547996521,
+      "learning_rate": 3.987192750660719e-06,
+      "loss": 0.0858,
+      "step": 50
+    },
+    {
+      "epoch": 1.0666666666666667,
+      "grad_norm": 2.225501775741577,
+      "learning_rate": 3.587417902020876e-06,
+      "loss": 0.109,
+      "step": 60
+    },
+    {
+      "epoch": 1.2444444444444445,
+      "grad_norm": 3.221173048019409,
+      "learning_rate": 3.1496829497545268e-06,
+      "loss": 0.077,
+      "step": 70
+    },
+    {
+      "epoch": 1.4222222222222223,
+      "grad_norm": 3.5979526042938232,
+      "learning_rate": 2.6892685546987724e-06,
+      "loss": 0.0734,
+      "step": 80
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 2.8982393741607666,
+      "learning_rate": 2.2222470825144806e-06,
+      "loss": 0.0877,
+      "step": 90
+    },
+    {
+      "epoch": 1.7777777777777777,
+      "grad_norm": 3.0191638469696045,
+      "learning_rate": 1.7649215418673847e-06,
+      "loss": 0.0499,
+      "step": 100
+    },
+    {
+      "epoch": 1.9555555555555557,
+      "grad_norm": 2.951284646987915,
+      "learning_rate": 1.3332564712129845e-06,
+      "loss": 0.0887,
+      "step": 110
+    },
+    {
+      "epoch": 2.1333333333333333,
+      "grad_norm": 2.143000602722168,
+      "learning_rate": 9.423206410612498e-07,
+      "loss": 0.0646,
+      "step": 120
+    },
+    {
+      "epoch": 2.311111111111111,
+      "grad_norm": 1.9689100980758667,
+      "learning_rate": 6.057610261367044e-07,
+      "loss": 0.0734,
+      "step": 130
+    },
+    {
+      "epoch": 2.488888888888889,
+      "grad_norm": 4.3230180740356445,
+      "learning_rate": 3.3532641026504415e-07,
+      "loss": 0.0898,
+      "step": 140
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 1.9908875226974487,
+      "learning_rate": 1.4045725421448332e-07,
+      "loss": 0.0622,
+      "step": 150
+    },
+    {
+      "epoch": 2.8444444444444446,
+      "grad_norm": 3.6597988605499268,
+      "learning_rate": 2.7956143581177874e-08,
+      "loss": 0.0601,
+      "step": 160
+    },
+    {
+      "epoch": 2.986666666666667,
+      "step": 168,
+      "total_flos": 1.6019812625547264e+16,
+      "train_loss": 0.12199947628236953,
+      "train_runtime": 363.0569,
+      "train_samples_per_second": 7.437,
+      "train_steps_per_second": 0.463
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 168,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "total_flos": 1.6019812625547264e+16,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}