turkish-gpt2-mini-M2-cleaned-politics720k-20ep
This model is a fine-tuned version of eminAydin/turkish-gpt2-mini-M0-cleaned-wiki480k-20ep on the None dataset. It achieves the following results on the evaluation set:
- Loss: 4.1592
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: 5e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.03
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
4.9645 | 1.0 | 9012 | 4.7521 |
4.4568 | 2.0 | 18025 | 4.5201 |
4.2558 | 3.0 | 27037 | 4.3970 |
4.1194 | 4.0 | 36050 | 4.3101 |
4.0163 | 5.0 | 45063 | 4.2547 |
3.934 | 6.0 | 54075 | 4.2125 |
3.8687 | 7.0 | 63088 | 4.1856 |
3.8202 | 8.0 | 72101 | 4.1686 |
3.7887 | 9.0 | 81113 | 4.1601 |
3.7726 | 10.0 | 90120 | 4.1592 |
Framework versions
- Transformers 4.39.3
- Pytorch 2.1.2
- Datasets 2.18.0
- Tokenizers 0.15.2
- Downloads last month
- 60