m2m-amharic-synonym-tuned
This model is a fine-tuned version of facebook/m2m100_418M on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.4384
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: 2.4863944860389072e-05
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- 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
- num_epochs: 43
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 4.5713 | 1.0 | 76 | 4.0324 |
| 1.5739 | 2.0 | 152 | 1.3050 |
| 0.557 | 3.0 | 228 | 0.5062 |
| 0.4377 | 4.0 | 304 | 0.4535 |
| 0.3907 | 5.0 | 380 | 0.4421 |
| 0.3683 | 6.0 | 456 | 0.4352 |
| 0.3443 | 7.0 | 532 | 0.4399 |
| 0.3339 | 8.0 | 608 | 0.4384 |
Framework versions
- Transformers 4.47.0
- Pytorch 2.5.1+cu121
- Datasets 3.2.0
- Tokenizers 0.21.0
- Downloads last month
- 5
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
🙋
Ask for provider support
Model tree for Beck90/m2m-amharic-synonym-tuned
Base model
facebook/m2m100_418M