model
This model was trained from scratch on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 2.4430
- Bleu: 11.7489
- Gen Len: 31.5264
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: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Use OptimizerNames.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: 1
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|---|
| 2.8232 | 0.0894 | 5000 | 2.6778 | 9.2453 | 33.6511 |
| 2.7385 | 0.1788 | 10000 | 2.6049 | 9.9439 | 32.4017 |
| 2.6999 | 0.2681 | 15000 | 2.5599 | 9.553 | 33.2342 |
| 2.6424 | 0.3575 | 20000 | 2.5284 | 10.1134 | 32.6427 |
| 2.6156 | 0.4469 | 25000 | 2.5038 | 10.6302 | 32.3189 |
| 2.5972 | 0.5363 | 30000 | 2.4853 | 11.5421 | 31.3456 |
| 2.5682 | 0.6256 | 35000 | 2.4688 | 11.3178 | 31.7832 |
| 2.5809 | 0.7150 | 40000 | 2.4590 | 11.6012 | 31.0891 |
| 2.5643 | 0.8044 | 45000 | 2.4506 | 11.7468 | 32.0279 |
| 2.5564 | 0.8938 | 50000 | 2.4451 | 11.7078 | 32.0276 |
| 2.5643 | 0.9831 | 55000 | 2.4430 | 11.7489 | 31.5264 |
Framework versions
- Transformers 4.48.1
- Pytorch 2.6.0+cu118
- Datasets 3.2.0
- Tokenizers 0.21.0
