m2m100_1.2B-ft-cy-to-en-2510
This model is a fine-tuned version of facebook/m2m100_1.2B on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.6758
- Bleu: 58.3947
- Gen Len: 24.7949
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: 1e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 6000
- training_steps: 30000
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|---|---|---|---|---|---|
| 1.4205 | 0.0166 | 2000 | 1.2444 | 39.2533 | 24.9409 |
| 1.1584 | 0.0332 | 4000 | 1.0151 | 46.658 | 24.9177 |
| 1.0218 | 0.0499 | 6000 | 0.9055 | 50.5177 | 24.7381 |
| 0.9319 | 0.0665 | 8000 | 0.8424 | 52.4629 | 24.673 |
| 0.8824 | 0.0831 | 10000 | 0.8024 | 53.6786 | 24.7644 |
| 0.8524 | 0.0997 | 12000 | 0.7723 | 54.9932 | 24.819 |
| 0.8225 | 0.1164 | 14000 | 0.7514 | 55.6039 | 24.8895 |
| 0.8103 | 0.1330 | 16000 | 0.7321 | 56.6636 | 24.8258 |
| 0.7891 | 0.1496 | 18000 | 0.7187 | 56.8648 | 24.7212 |
| 0.7755 | 0.1662 | 20000 | 0.7078 | 57.3337 | 24.8745 |
| 0.7741 | 0.1829 | 22000 | 0.6976 | 57.7373 | 24.7467 |
| 0.7604 | 0.1995 | 24000 | 0.6878 | 57.7788 | 24.8095 |
| 0.7453 | 0.2161 | 26000 | 0.6811 | 58.098 | 24.8304 |
| 0.752 | 0.2327 | 28000 | 0.6781 | 58.408 | 24.8108 |
| 0.7412 | 0.2494 | 30000 | 0.6758 | 58.3947 | 24.7949 |
Framework versions
- Transformers 4.56.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.22.1