metadata
license: apache-2.0
base_model: Karzan/gpt2-walamakan
tags:
- generated_from_trainer
model-index:
- name: gpt2-walamakan-2
results: []
gpt2-walamakan-2
This model is a fine-tuned version of Karzan/gpt2-walamakan on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 6.7392
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: 3e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 30
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.523 | 1.0 | 94 | 6.4968 |
| 0.5212 | 2.0 | 188 | 6.4501 |
| 0.488 | 3.0 | 282 | 6.4814 |
| 0.4723 | 4.0 | 376 | 6.5004 |
| 0.4452 | 5.0 | 470 | 6.5328 |
| 0.4442 | 6.0 | 564 | 6.5507 |
| 0.4147 | 7.0 | 658 | 6.5598 |
| 0.397 | 8.0 | 752 | 6.5623 |
| 0.3868 | 9.0 | 846 | 6.5642 |
| 0.3686 | 10.0 | 940 | 6.5713 |
| 0.3553 | 11.0 | 1034 | 6.6027 |
| 0.338 | 12.0 | 1128 | 6.5953 |
| 0.3344 | 13.0 | 1222 | 6.6386 |
| 0.315 | 14.0 | 1316 | 6.6202 |
| 0.3096 | 15.0 | 1410 | 6.6239 |
| 0.2961 | 16.0 | 1504 | 6.6648 |
| 0.2899 | 17.0 | 1598 | 6.6663 |
| 0.2782 | 18.0 | 1692 | 6.6750 |
| 0.2642 | 19.0 | 1786 | 6.6777 |
| 0.2541 | 20.0 | 1880 | 6.6807 |
| 0.2502 | 21.0 | 1974 | 6.6956 |
| 0.2453 | 22.0 | 2068 | 6.7099 |
| 0.2485 | 23.0 | 2162 | 6.7159 |
| 0.2342 | 24.0 | 2256 | 6.7149 |
| 0.2226 | 25.0 | 2350 | 6.7288 |
| 0.22 | 26.0 | 2444 | 6.7302 |
| 0.2172 | 27.0 | 2538 | 6.7315 |
| 0.2185 | 28.0 | 2632 | 6.7374 |
| 0.2131 | 29.0 | 2726 | 6.7361 |
| 0.2089 | 30.0 | 2820 | 6.7392 |
Framework versions
- Transformers 4.32.0
- Pytorch 2.0.1+cu118
- Datasets 2.14.4
- Tokenizers 0.13.3