<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
nllb-200-distilled-600M-en-to-dz
This model is a fine-tuned version of facebook/nllb-200-distilled-600M on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.4130
- Bleu: 3.9043
- Gen Len: 16.5447
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|---|---|---|---|---|---|
| 1.7451 | 1.0 | 1688 | 1.5913 | 2.285 | 17.3937 |
| 1.5595 | 2.0 | 3376 | 1.4808 | 1.9817 | 16.787 |
| 1.429 | 3.0 | 5064 | 1.4367 | 2.7438 | 16.802 |
| 1.3653 | 4.0 | 6752 | 1.4180 | 4.3181 | 16.5257 |
| 1.3291 | 5.0 | 8440 | 1.4130 | 3.9043 | 16.6407 |
Framework versions
- Transformers 4.29.0.dev0
- Pytorch 2.0.0+cu118
- Datasets 2.11.0
- Tokenizers 0.13.3
