Model save

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/nllb-200-distilled-600M](https://huggingface.co/facebook/nllb-200-distilled-600M) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 8.0911
+- Loss: 7.1958
 
 ## Model description
 
@@ -35,8 +35,8 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 2e-05
-- train_batch_size: 256
-- eval_batch_size: 256
+- train_batch_size: 1024
+- eval_batch_size: 512
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
@@ -47,16 +47,16 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
-| No log        | 1.0   | 1    | 8.2732          |
-| No log        | 2.0   | 2    | 8.1703          |
-| No log        | 3.0   | 3    | 8.0670          |
-| No log        | 4.0   | 4    | 8.0670          |
-| No log        | 5.0   | 5    | 8.0239          |
-| No log        | 6.0   | 6    | 8.0299          |
-| No log        | 7.0   | 7    | 8.0546          |
-| No log        | 8.0   | 8    | 8.0734          |
-| No log        | 9.0   | 9    | 8.0845          |
-| 7.3874        | 10.0  | 10   | 8.0911          |
+| No log        | 1.0   | 1    | 7.3299          |
+| No log        | 2.0   | 2    | 7.3299          |
+| No log        | 3.0   | 3    | 7.2664          |
+| No log        | 4.0   | 4    | 7.2299          |
+| No log        | 5.0   | 5    | 7.1992          |
+| No log        | 6.0   | 6    | 7.1871          |
+| No log        | 7.0   | 7    | 7.1885          |
+| No log        | 8.0   | 8    | 7.1922          |
+| No log        | 9.0   | 9    | 7.1944          |
+| 7.3798        | 10.0  | 10   | 7.1958          |
 
 
 ### Framework versions