End of training

README.md

@@ -1,7 +1,7 @@
 ---
 library_name: peft
 license: llama3.2
-base_model: meta-llama/Llama-3.2-1B
+base_model: meta-llama/Llama-3.2-3B
 tags:
 - generated_from_trainer
 metrics:
@@ -17,11 +17,11 @@ should probably proofread and complete it, then remove this comment. -->
 
 # mrpc-lora
 
-This model is a fine-tuned version of [meta-llama/Llama-3.2-1B](https://huggingface.co/meta-llama/Llama-3.2-1B) on an unknown dataset.
+This model is a fine-tuned version of [meta-llama/Llama-3.2-3B](https://huggingface.co/meta-llama/Llama-3.2-3B) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.8326
-- Accuracy: 0.8529
-- F1: 0.8921
+- Loss: 0.4694
+- Accuracy: 0.7917
+- F1: 0.8537
 
 ## Model description
 
@@ -41,28 +41,23 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 2e-05
-- train_batch_size: 8
-- eval_batch_size: 8
+- train_batch_size: 16
+- eval_batch_size: 16
 - seed: 42
 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: linear
-- num_epochs: 10
+- num_epochs: 5
 - mixed_precision_training: Native AMP
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1     |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|
-| 0.6131        | 1.0   | 459  | 0.4045          | 0.8186   | 0.8746 |
-| 0.3038        | 2.0   | 918  | 0.3880          | 0.8578   | 0.8993 |
-| 0.2163        | 3.0   | 1377 | 0.4428          | 0.8505   | 0.8946 |
-| 0.2105        | 4.0   | 1836 | 0.5322          | 0.8529   | 0.8958 |
-| 0.4452        | 5.0   | 2295 | 0.6446          | 0.8407   | 0.8799 |
-| 0.3287        | 6.0   | 2754 | 0.6270          | 0.8554   | 0.8974 |
-| 0.2136        | 7.0   | 3213 | 0.7087          | 0.8456   | 0.8897 |
-| 0.1533        | 8.0   | 3672 | 0.7688          | 0.8407   | 0.8825 |
-| 0.1739        | 9.0   | 4131 | 0.8213          | 0.8480   | 0.8920 |
-| 0.2349        | 10.0  | 4590 | 0.8326          | 0.8529   | 0.8921 |
+| 0.6611        | 1.0   | 230  | 0.6236          | 0.6642   | 0.7720 |
+| 0.5343        | 2.0   | 460  | 0.5438          | 0.7255   | 0.8069 |
+| 0.4298        | 3.0   | 690  | 0.5006          | 0.7745   | 0.8414 |
+| 0.4538        | 4.0   | 920  | 0.4770          | 0.7917   | 0.8557 |
+| 0.4486        | 5.0   | 1150 | 0.4694          | 0.7917   | 0.8537 |
 
 
 ### Framework versions