nllb-indo-en
This model is a fine-tuned version of facebook/nllb-200-distilled-600M on Fleurs Dataset. It achieves the following results on the evaluation set:
- Loss: 0.1114
Intended uses & limitations
Note: This model is still under investigation due to overfitting on validation set.
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 16
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- training_steps: 1000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 53.2469 | 1.0 | 178 | 2.3718 |
| 8.4186 | 2.0 | 356 | 0.1943 |
| 0.9752 | 3.0 | 534 | 0.1254 |
| 0.5674 | 4.0 | 712 | 0.1147 |
| 0.4423 | 5.0 | 890 | 0.1109 |
| 0.3477 | 5.6180 | 1000 | 0.1114 |
Framework versions
- Transformers 4.47.1
- Pytorch 2.5.1+cu121
- Datasets 3.2.0
- Tokenizers 0.21.0
- Downloads last month
- 1,169
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.
Model tree for cobrayyxx/nllb-indo-en-dum
Base model
facebook/nllb-200-distilled-600M