End of training

README.md

@@ -4,6 +4,7 @@ library_name: peft
 license: apache-2.0
 tags:
 - llama-factory
+- lora
 - generated_from_trainer
 model-index:
 - name: Mistral-Nemo-12B-Instruct-SFT-Half
@@ -15,7 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # Mistral-Nemo-12B-Instruct-SFT-Half
 
-This model is a fine-tuned version of [mistralai/Mistral-Nemo-Instruct-2407](https://huggingface.co/mistralai/Mistral-Nemo-Instruct-2407) on an unknown dataset.
+This model is a fine-tuned version of [mistralai/Mistral-Nemo-Instruct-2407](https://huggingface.co/mistralai/Mistral-Nemo-Instruct-2407) on the bct_non_cot_sft_500 dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.4380
 
 ## Model description
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 2.986666666666667,
+    "eval_loss": 0.43796446919441223,
+    "eval_runtime": 2.1916,
+    "eval_samples_per_second": 22.814,
+    "eval_steps_per_second": 11.407,
+    "total_flos": 1.134821632868352e+16,
+    "train_loss": 0.7413791517416636,
+    "train_runtime": 233.6297,
+    "train_samples_per_second": 5.778,
+    "train_steps_per_second": 0.36
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 2.986666666666667,
+    "eval_loss": 0.43796446919441223,
+    "eval_runtime": 2.1916,
+    "eval_samples_per_second": 22.814,
+    "eval_steps_per_second": 11.407
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.986666666666667,
+    "total_flos": 1.134821632868352e+16,
+    "train_loss": 0.7413791517416636,
+    "train_runtime": 233.6297,
+    "train_samples_per_second": 5.778,
+    "train_steps_per_second": 0.36
+}

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": 3.758638381958008,
+      "learning_rate": 4.444444444444444e-06,
+      "loss": 1.2892,
+      "step": 10
+    },
+    {
+      "epoch": 0.7111111111111111,
+      "grad_norm": 3.2213594913482666,
+      "learning_rate": 4.824441214720629e-06,
+      "loss": 1.0878,
+      "step": 20
+    },
+    {
+      "epoch": 1.0666666666666667,
+      "grad_norm": 2.8044538497924805,
+      "learning_rate": 4.249158351283414e-06,
+      "loss": 0.8957,
+      "step": 30
+    },
+    {
+      "epoch": 1.4222222222222223,
+      "grad_norm": 2.3454504013061523,
+      "learning_rate": 3.3714301183045382e-06,
+      "loss": 0.7399,
+      "step": 40
+    },
+    {
+      "epoch": 1.7777777777777777,
+      "grad_norm": 2.036797285079956,
+      "learning_rate": 2.3430237011767166e-06,
+      "loss": 0.6107,
+      "step": 50
+    },
+    {
+      "epoch": 2.1333333333333333,
+      "grad_norm": 1.5367364883422852,
+      "learning_rate": 1.3417599122003464e-06,
+      "loss": 0.4844,
+      "step": 60
+    },
+    {
+      "epoch": 2.488888888888889,
+      "grad_norm": 2.367539644241333,
+      "learning_rate": 5.407663566854008e-07,
+      "loss": 0.4673,
+      "step": 70
+    },
+    {
+      "epoch": 2.8444444444444446,
+      "grad_norm": 2.4404282569885254,
+      "learning_rate": 7.854209717842231e-08,
+      "loss": 0.4532,
+      "step": 80
+    },
+    {
+      "epoch": 2.986666666666667,
+      "step": 84,
+      "total_flos": 1.134821632868352e+16,
+      "train_loss": 0.7413791517416636,
+      "train_runtime": 233.6297,
+      "train_samples_per_second": 5.778,
+      "train_steps_per_second": 0.36
+    }
+  ],
+  "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": 1.134821632868352e+16,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}