metadata
library_name: peft
base_model: roneneldan/TinyStories-1M
tags:
- generated_from_trainer
model-index:
- name: test_1M_1-2025-02-20-18-21
results: []
test_1M_1-2025-02-20-18-21
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: 2.4115
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: 6
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.5947 | 0.5 | 297 | 2.5063 |
| 2.5953 | 1.0 | 594 | 2.4885 |
| 2.4797 | 1.5 | 891 | 2.4663 |
| 2.3619 | 2.0 | 1188 | 2.4567 |
| 2.4461 | 2.5 | 1485 | 2.4477 |
| 2.6288 | 3.0 | 1782 | 2.4462 |
| 2.4072 | 3.5 | 2079 | 2.4369 |
| 2.417 | 4.0 | 2376 | 2.4303 |
| 2.3747 | 4.5 | 2673 | 2.4293 |
| 2.313 | 5.0 | 2970 | 2.4210 |
| 2.3675 | 5.5 | 3267 | 2.4159 |
| 2.402 | 6.0 | 3564 | 2.4115 |
Framework versions
- PEFT 0.14.0
- Transformers 4.48.1
- Pytorch 2.5.1+cu124
- Datasets 3.2.0
- Tokenizers 0.21.0