metadata
library_name: peft
license: mit
base_model: gpt2
tags:
- generated_from_trainer
model-index:
- name: Se124M500KInfPrompt_endtoken
results: []
Se124M500KInfPrompt_endtoken
This model is a fine-tuned version of gpt2 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.6716
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: 5e-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: 50
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.1898 | 1.0 | 5427 | 0.7433 |
| 0.1857 | 2.0 | 10854 | 0.7238 |
| 0.1843 | 3.0 | 16281 | 0.7118 |
| 0.1813 | 4.0 | 21708 | 0.7045 |
| 0.1802 | 5.0 | 27135 | 0.6990 |
| 0.1785 | 6.0 | 32562 | 0.6944 |
| 0.1769 | 7.0 | 37989 | 0.6918 |
| 0.1743 | 8.0 | 43416 | 0.6875 |
| 0.1752 | 9.0 | 48843 | 0.6854 |
| 0.1756 | 10.0 | 54270 | 0.6854 |
| 0.1736 | 11.0 | 59697 | 0.6837 |
| 0.1756 | 12.0 | 65124 | 0.6812 |
| 0.173 | 13.0 | 70551 | 0.6798 |
| 0.1737 | 14.0 | 75978 | 0.6791 |
| 0.1741 | 15.0 | 81405 | 0.6783 |
| 0.177 | 16.0 | 86832 | 0.6771 |
| 0.1734 | 17.0 | 92259 | 0.6765 |
| 0.1719 | 18.0 | 97686 | 0.6760 |
| 0.1737 | 19.0 | 103113 | 0.6763 |
| 0.1716 | 20.0 | 108540 | 0.6747 |
| 0.1713 | 21.0 | 113967 | 0.6741 |
| 0.1739 | 22.0 | 119394 | 0.6738 |
| 0.1694 | 23.0 | 124821 | 0.6737 |
| 0.1703 | 24.0 | 130248 | 0.6743 |
| 0.1697 | 25.0 | 135675 | 0.6730 |
| 0.172 | 26.0 | 141102 | 0.6731 |
| 0.1711 | 27.0 | 146529 | 0.6720 |
| 0.1726 | 28.0 | 151956 | 0.6720 |
| 0.1703 | 29.0 | 157383 | 0.6716 |
| 0.1732 | 30.0 | 162810 | 0.6716 |
| 0.171 | 31.0 | 168237 | 0.6719 |
Framework versions
- PEFT 0.15.1
- Transformers 4.51.3
- Pytorch 2.6.0+cu118
- Datasets 3.5.0
- Tokenizers 0.21.1