opt-125m-finetuned-wikitext-103
This model is a fine-tuned version of facebook/opt-125m on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 3.3132
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: 8
- eval_batch_size: 8
- 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: 10
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.3229 | 1.0 | 638 | 3.0664 |
| 2.2408 | 2.0 | 1276 | 3.1068 |
| 2.1741 | 3.0 | 1914 | 3.1402 |
| 2.0303 | 4.0 | 2552 | 3.1786 |
| 1.957 | 5.0 | 3190 | 3.2170 |
| 1.9129 | 6.0 | 3828 | 3.2481 |
| 1.8651 | 7.0 | 4466 | 3.2738 |
| 1.7848 | 8.0 | 5104 | 3.2900 |
| 1.7577 | 9.0 | 5742 | 3.3033 |
| 1.732 | 10.0 | 6380 | 3.3132 |
Framework versions
- Transformers 4.51.3
- Pytorch 2.6.0+cu124
- Datasets 3.5.1
- Tokenizers 0.21.1
- Downloads last month
- 4
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
🙋
Ask for provider support
Model tree for quansuv/opt-125m-finetuned-wikitext-103
Base model
facebook/opt-125m