End of training

README.md

@@ -18,8 +18,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [IAmSkyDra/BARTBana_v5](https://huggingface.co/IAmSkyDra/BARTBana_v5) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.8556
-- Sacrebleu: 1.5240
+- Loss: 2.7448
+- Sacrebleu: 1.3613
 
 ## Model description
 
@@ -39,10 +39,10 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 2e-05
-- train_batch_size: 64
-- eval_batch_size: 64
+- train_batch_size: 128
+- eval_batch_size: 128
 - seed: 42
-- optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: linear
 - num_epochs: 3
 - mixed_precision_training: Native AMP
@@ -51,14 +51,14 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step  | Validation Loss | Sacrebleu |
 |:-------------:|:-----:|:-----:|:---------------:|:---------:|
-| 0.3694        | 1.0   | 15848 | 2.5203          | 1.0776    |
-| 0.2075        | 2.0   | 31696 | 2.7832          | 1.3857    |
-| 0.1651        | 3.0   | 47544 | 2.8556          | 1.5240    |
+| 0.4874        | 1.0   | 8322  | 2.4353          | 0.9260    |
+| 0.2892        | 2.0   | 16644 | 2.6754          | 1.2124    |
+| 0.2431        | 3.0   | 24966 | 2.7448          | 1.3613    |
 
 
 ### Framework versions
 
 - Transformers 4.57.3
-- Pytorch 2.9.1+cu128
-- Datasets 4.4.2
+- Pytorch 2.9.0+cu126
+- Datasets 4.0.0
 - Tokenizers 0.22.1