metadata
library_name: peft
base_model: roneneldan/TinyStories-1M
tags:
- generated_from_trainer
model-index:
- name: test_1M_1-2025-02-16-18-59
results: []
test_1M_1-2025-02-16-18-59
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: 2.3658
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: 30
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.5392 | 0.5 | 297 | 2.4605 |
| 2.4817 | 1.0 | 594 | 2.4445 |
| 2.4207 | 1.5 | 891 | 2.4344 |
| 2.4092 | 2.0 | 1188 | 2.4328 |
| 2.4385 | 2.5 | 1485 | 2.4275 |
| 2.5104 | 3.0 | 1782 | 2.4149 |
| 2.3552 | 3.5 | 2079 | 2.4131 |
| 2.402 | 4.0 | 2376 | 2.4120 |
| 2.4328 | 4.5 | 2673 | 2.4143 |
| 2.4508 | 5.0 | 2970 | 2.4052 |
| 2.2452 | 5.5 | 3267 | 2.4064 |
| 2.5212 | 6.0 | 3564 | 2.4137 |
| 2.3123 | 6.5 | 3861 | 2.4038 |
| 2.3935 | 7.0 | 4158 | 2.4001 |
| 2.2864 | 7.5 | 4455 | 2.3967 |
| 2.3657 | 8.0 | 4752 | 2.3980 |
| 2.5036 | 8.5 | 5049 | 2.4018 |
| 2.3336 | 9.0 | 5346 | 2.3965 |
| 2.3799 | 9.5 | 5643 | 2.3916 |
| 2.478 | 10.0 | 5940 | 2.3979 |
| 2.3376 | 10.5 | 6237 | 2.3923 |
| 2.3039 | 11.0 | 6534 | 2.3923 |
| 2.3658 | 11.5 | 6831 | 2.3900 |
| 2.473 | 12.0 | 7128 | 2.3901 |
| 2.3923 | 12.5 | 7425 | 2.3869 |
| 2.4122 | 13.0 | 7722 | 2.3867 |
| 2.4238 | 13.5 | 8019 | 2.3870 |
| 2.4234 | 14.0 | 8316 | 2.3843 |
| 2.4062 | 14.5 | 8613 | 2.3869 |
| 2.3188 | 15.0 | 8910 | 2.3813 |
| 2.2888 | 15.5 | 9207 | 2.3835 |
| 2.3326 | 16.0 | 9504 | 2.3779 |
| 2.3273 | 16.5 | 9801 | 2.3807 |
| 2.3338 | 17.0 | 10098 | 2.3788 |
| 2.4337 | 17.5 | 10395 | 2.3792 |
| 2.3396 | 18.0 | 10692 | 2.3800 |
| 2.3172 | 18.5 | 10989 | 2.3806 |
| 2.3586 | 19.0 | 11286 | 2.3807 |
| 2.3708 | 19.5 | 11583 | 2.3789 |
| 2.449 | 20.0 | 11880 | 2.3762 |
| 2.3071 | 20.5 | 12177 | 2.3786 |
| 2.2589 | 21.0 | 12474 | 2.3750 |
| 2.2423 | 21.5 | 12771 | 2.3749 |
| 2.2852 | 22.0 | 13068 | 2.3737 |
| 2.2754 | 22.5 | 13365 | 2.3750 |
| 2.2977 | 23.0 | 13662 | 2.3737 |
| 2.2701 | 23.5 | 13959 | 2.3701 |
| 2.2638 | 24.0 | 14256 | 2.3726 |
| 2.377 | 24.5 | 14553 | 2.3733 |
| 2.3774 | 25.0 | 14850 | 2.3725 |
| 2.2137 | 25.5 | 15147 | 2.3722 |
| 2.3267 | 26.0 | 15444 | 2.3681 |
| 2.2415 | 26.5 | 15741 | 2.3706 |
| 2.2957 | 27.0 | 16038 | 2.3687 |
| 2.3003 | 27.5 | 16335 | 2.3678 |
| 2.3662 | 28.0 | 16632 | 2.3678 |
| 2.305 | 28.5 | 16929 | 2.3673 |
| 2.2603 | 29.0 | 17226 | 2.3667 |
| 2.2806 | 29.5 | 17523 | 2.3665 |
| 2.2674 | 30.0 | 17820 | 2.3658 |
Framework versions
- PEFT 0.14.0
- Transformers 4.48.1
- Pytorch 2.5.1+cu124
- Datasets 3.2.0
- Tokenizers 0.21.0