gpt_16_4_3e-5_lp5_nb5
This model is a fine-tuned version of skt/kogpt2-base-v2 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 3.8872
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: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 4.0
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 3.311 | 0.38 | 1000 | 4.0339 |
| 3.1133 | 0.76 | 2000 | 3.9777 |
| 2.9875 | 1.13 | 3000 | 3.9546 |
| 2.8697 | 1.51 | 4000 | 3.9269 |
| 2.8669 | 1.89 | 5000 | 3.9159 |
| 2.7308 | 2.27 | 6000 | 3.9066 |
| 2.709 | 2.64 | 7000 | 3.8995 |
| 2.6979 | 3.02 | 8000 | 3.8976 |
| 2.5878 | 3.4 | 9000 | 3.8978 |
| 2.5824 | 3.78 | 10000 | 3.8872 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.9.0+cu102
- Datasets 2.8.0
- Tokenizers 0.13.2