| --- |
| library_name: transformers |
| tags: |
| - generated_from_trainer |
| metrics: |
| - accuracy |
| - bleu |
| model-index: |
| - name: gpt2-wikitext |
| results: [] |
| --- |
| |
| <!-- This model card has been generated automatically according to the information the Trainer had access to. You |
| should probably proofread and complete it, then remove this comment. --> |
|
|
| # gpt2-wikitext |
|
|
| This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset. |
| It achieves the following results on the evaluation set: |
| - Loss: 3.1673 |
| - Accuracy: 0.4215 |
| - Perplexity: 23.7437 |
| - Bleu: 0.1486 |
|
|
| ## 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: 0.0001 |
| - train_batch_size: 64 |
| - eval_batch_size: 64 |
| - 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 |
| - lr_scheduler_warmup_ratio: 0.1 |
| - num_epochs: 5 |
| |
| ### Training results |
| |
| | Training Loss | Epoch | Step | Validation Loss | Accuracy | Perplexity | Bleu | |
| |:-------------:|:------:|:----:|:---------------:|:--------:|:----------:|:------:| |
| | 6.0809 | 0.2806 | 500 | 5.9591 | 0.1879 | 387.2722 | 0.0323 | |
| | 5.0659 | 0.5612 | 1000 | 4.9189 | 0.2618 | 136.8551 | 0.0639 | |
| | 4.338 | 0.8418 | 1500 | 4.2174 | 0.3219 | 67.8594 | 0.0857 | |
| | 3.9468 | 1.1223 | 2000 | 3.8832 | 0.3532 | 48.5783 | 0.1094 | |
| | 3.7577 | 1.4029 | 2500 | 3.7060 | 0.3685 | 40.6890 | 0.1226 | |
| | 3.6483 | 1.6835 | 3000 | 3.5831 | 0.3787 | 35.9858 | 0.1296 | |
| | 3.5432 | 1.9641 | 3500 | 3.4948 | 0.3875 | 32.9448 | 0.1360 | |
| | 3.4221 | 2.2447 | 4000 | 3.4280 | 0.3939 | 30.8160 | 0.1306 | |
| | 3.3602 | 2.5253 | 4500 | 3.3724 | 0.3991 | 29.1478 | 0.1391 | |
| | 3.3285 | 2.8058 | 5000 | 3.3261 | 0.4038 | 27.8284 | 0.1369 | |
| | 3.2072 | 3.0864 | 5500 | 3.2882 | 0.4077 | 26.7936 | 0.1447 | |
| | 3.2002 | 3.3670 | 6000 | 3.2611 | 0.4112 | 26.0792 | 0.1472 | |
| | 3.1782 | 3.6476 | 6500 | 3.2317 | 0.4138 | 25.3223 | 0.1421 | |
| | 3.153 | 3.9282 | 7000 | 3.2080 | 0.4164 | 24.7294 | 0.1415 | |
| | 3.0846 | 4.2088 | 7500 | 3.1915 | 0.4185 | 24.3249 | 0.1470 | |
| | 3.0469 | 4.4893 | 8000 | 3.1789 | 0.4199 | 24.0205 | 0.1444 | |
| | 3.0567 | 4.7699 | 8500 | 3.1673 | 0.4215 | 23.7437 | 0.1486 | |
| |
| |
| ### Framework versions |
| |
| - Transformers 4.49.0 |
| - Pytorch 2.6.0+cu124 |
| - Datasets 3.3.2 |
| - Tokenizers 0.21.0 |
| |
