mbart-large-50-b16-e16-lr2e-05-126k-jupyter
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.1198
- Bleu: 84.233
- Gen Len Ratio: 1.0012
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: Use OptimizerNames.ADAFACTOR and the args are:
No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 16
Training results
| Training Loss | Epoch | Step | Bleu | Gen Len Ratio | Validation Loss |
|---|
| 0.1532 | 1.7769 | 14000 | 84.233 | 1.0012 | 0.1198 |
| 0.0485 | 3.5538 | 28000 | 72.1008 | 0.8848 | 0.1352 |
| 0.0196 | 5.3306 | 42000 | 64.2991 | 0.8081 | 0.1477 |
| 0.0083 | 7.1075 | 56000 | 56.8508 | 0.7444 | 0.1608 |
| 0.0031 | 8.8844 | 70000 | 63.6676 | 0.7982 | 0.1675 |
| 0.0017 | 10.6613 | 84000 | 72.8835 | 0.8861 | 0.1744 |
| 0.0008 | 12.4381 | 98000 | 72.2447 | 0.8851 | 0.1819 |
| 0.0005 | 14.2150 | 112000 | 69.4106 | 0.8549 | 0.1856 |
| 0.0003 | 15.9919 | 126000 | 67.0096 | 0.8301 | 0.1877 |
Framework versions
- Transformers 4.49.0
- Pytorch 2.6.0+cu124
- Datasets 3.4.1
- Tokenizers 0.21.1
