Added Noise to english source, fine-tuned with denoising, version mbart_endenoised_v2

README.md

@@ -15,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [facebook/mbart-large-50-one-to-many-mmt](https://huggingface.co/facebook/mbart-large-50-one-to-many-mmt) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.3923
+- Loss: 0.3740
 
 ## Model description
 
@@ -46,13 +46,13 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
-| 0.4815        | 1.0   | 625  | 0.3941          |
-| 0.3489        | 2.0   | 1250 | 0.3923          |
+| 0.444         | 1.0   | 1875 | 0.3807          |
+| 0.3206        | 2.0   | 3750 | 0.3740          |
 
 
 ### Framework versions
 
 - Transformers 4.47.0
 - Pytorch 2.5.1+cu121
-- Datasets 3.1.0
+- Datasets 3.2.0
 - Tokenizers 0.21.0