Model save

README.md

@@ -19,7 +19,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [bigcode/starcoder2-3b](https://huggingface.co/bigcode/starcoder2-3b) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.4791
+- Loss: 0.4398
 
 ## Model description
 
@@ -38,21 +38,24 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 3e-05
-- train_batch_size: 2
-- eval_batch_size: 2
+- learning_rate: 1e-05
+- train_batch_size: 1
+- eval_batch_size: 1
 - seed: 42
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 2
 - 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: 2
+- num_epochs: 3
 - mixed_precision_training: Native AMP
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
-| 1.9751        | 1.0   | 60   | 0.4862          |
-| 1.4621        | 2.0   | 120  | 0.4791          |
+| 2.574         | 1.0   | 60   | 0.4433          |
+| 2.5626        | 2.0   | 120  | 0.4411          |
+| 3.1912        | 3.0   | 180  | 0.4398          |
 
 
 ### Framework versions