metadata
license: mit
base_model: gpt2
tags:
- generated_from_trainer
model-index:
- name: polynomial_1450_7e-4_16b_w0.075
results: []
polynomial_1450_7e-4_16b_w0.075
This model is a fine-tuned version of gpt2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 3.0229
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: 0.0007
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 10
- total_train_batch_size: 160
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: polynomial
- lr_scheduler_warmup_steps: 250
- training_steps: 1450
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
9.0635 | 0.1029 | 50 | 7.2772 |
6.7176 | 0.2058 | 100 | 6.2557 |
6.0147 | 0.3088 | 150 | 5.7199 |
5.5539 | 0.4117 | 200 | 5.3464 |
5.2292 | 0.5146 | 250 | 5.0625 |
4.9374 | 0.6175 | 300 | 4.7728 |
4.6985 | 0.7205 | 350 | 4.5613 |
4.4993 | 0.8234 | 400 | 4.3770 |
4.3227 | 0.9263 | 450 | 4.1914 |
4.1342 | 1.0292 | 500 | 4.0022 |
3.8927 | 1.1322 | 550 | 3.8166 |
3.757 | 1.2351 | 600 | 3.6654 |
3.6277 | 1.3380 | 650 | 3.5614 |
3.5379 | 1.4409 | 700 | 3.4772 |
3.4642 | 1.5438 | 750 | 3.4076 |
3.395 | 1.6468 | 800 | 3.3542 |
3.3287 | 1.7497 | 850 | 3.3034 |
3.2872 | 1.8526 | 900 | 3.2609 |
3.2545 | 1.9555 | 950 | 3.2268 |
3.1229 | 2.0585 | 1000 | 3.1925 |
3.0573 | 2.1614 | 1050 | 3.1616 |
3.0339 | 2.2643 | 1100 | 3.1372 |
3.0204 | 2.3672 | 1150 | 3.1133 |
2.99 | 2.4702 | 1200 | 3.0949 |
2.9809 | 2.5731 | 1250 | 3.0720 |
2.9524 | 2.6760 | 1300 | 3.0536 |
2.9267 | 2.7789 | 1350 | 3.0392 |
2.9453 | 2.8818 | 1400 | 3.0295 |
2.93 | 2.9848 | 1450 | 3.0229 |
Framework versions
- Transformers 4.40.1
- Pytorch 2.3.0+cu121
- Datasets 2.19.0
- Tokenizers 0.19.1