metadata
library_name: peft
base_model: roneneldan/TinyStories-1M
tags:
- generated_from_trainer
model-index:
- name: test_1M_1-2025-02-13-16-57
results: []
test_1M_1-2025-02-13-16-57
This model is a fine-tuned version of roneneldan/TinyStories-1M on the None dataset. It achieves the following results on the evaluation set:
- Loss: 2.5458
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: 2.5e-05
- train_batch_size: 4
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 8
- optimizer: Use OptimizerNames.PAGED_ADAMW_8BIT with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 1
- num_epochs: 20
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.5623 | 0.5 | 297 | 2.6254 |
| 2.6957 | 1.0 | 594 | 2.6033 |
| 2.533 | 1.5 | 891 | 2.5928 |
| 2.583 | 2.0 | 1188 | 2.5871 |
| 2.4467 | 2.5 | 1485 | 2.5803 |
| 2.5157 | 3.0 | 1782 | 2.5742 |
| 2.5378 | 3.5 | 2079 | 2.5718 |
| 2.4285 | 4.0 | 2376 | 2.5732 |
| 2.4606 | 4.5 | 2673 | 2.5643 |
| 2.5373 | 5.0 | 2970 | 2.5660 |
| 2.5087 | 5.5 | 3267 | 2.5616 |
| 2.5158 | 6.0 | 3564 | 2.5620 |
| 2.4755 | 6.5 | 3861 | 2.5607 |
| 2.5964 | 7.0 | 4158 | 2.5595 |
| 2.54 | 7.5 | 4455 | 2.5608 |
| 2.5001 | 8.0 | 4752 | 2.5558 |
| 2.4152 | 8.5 | 5049 | 2.5562 |
| 2.5635 | 9.0 | 5346 | 2.5559 |
| 2.5722 | 9.5 | 5643 | 2.5550 |
| 2.5334 | 10.0 | 5940 | 2.5533 |
| 2.5433 | 10.5 | 6237 | 2.5540 |
| 2.4406 | 11.0 | 6534 | 2.5531 |
| 2.5045 | 11.5 | 6831 | 2.5528 |
| 2.5424 | 12.0 | 7128 | 2.5529 |
| 2.4114 | 12.5 | 7425 | 2.5508 |
| 2.4601 | 13.0 | 7722 | 2.5494 |
| 2.4751 | 13.5 | 8019 | 2.5519 |
| 2.4457 | 14.0 | 8316 | 2.5519 |
| 2.5539 | 14.5 | 8613 | 2.5500 |
| 2.6792 | 15.0 | 8910 | 2.5500 |
| 2.5712 | 15.5 | 9207 | 2.5498 |
| 2.4492 | 16.0 | 9504 | 2.5476 |
| 2.4512 | 16.5 | 9801 | 2.5481 |
| 2.417 | 17.0 | 10098 | 2.5482 |
| 2.5693 | 17.5 | 10395 | 2.5478 |
| 2.5358 | 18.0 | 10692 | 2.5482 |
| 2.4265 | 18.5 | 10989 | 2.5477 |
| 2.4858 | 19.0 | 11286 | 2.5475 |
| 2.4319 | 19.5 | 11583 | 2.5471 |
| 2.4089 | 20.0 | 11880 | 2.5458 |
Framework versions
- PEFT 0.14.0
- Transformers 4.48.1
- Pytorch 2.5.1+cu124
- Datasets 3.2.0
- Tokenizers 0.21.0