gpt2Kaggle3
This model is a fine-tuned version of ytu-ce-cosmos/turkish-gpt2 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 6.8194
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: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 128
- total_train_batch_size: 2048
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.03
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
9.3396 | 0.99 | 108 | 7.9615 |
7.8999 | 2.0 | 217 | 7.4396 |
7.4033 | 3.0 | 326 | 7.0468 |
7.1172 | 4.0 | 435 | 6.8595 |
6.9003 | 4.97 | 540 | 6.8194 |
Framework versions
- Transformers 4.39.3
- Pytorch 2.1.2
- Datasets 2.18.0
- Tokenizers 0.15.2
- Downloads last month
- 3