gpt-expt-sp-v3-3-mixed
This model is a fine-tuned version of gpt2 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0486
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: 0.0005
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 256
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 1000
- num_epochs: 300
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.4472 | 27.62 | 5000 | 0.0555 |
| 0.0593 | 55.25 | 10000 | 0.0511 |
| 0.054 | 82.87 | 15000 | 0.0502 |
| 0.0519 | 110.5 | 20000 | 0.0493 |
| 0.0508 | 138.12 | 25000 | 0.0490 |
| 0.05 | 165.74 | 30000 | 0.0488 |
| 0.0495 | 193.37 | 35000 | 0.0487 |
| 0.0492 | 220.99 | 40000 | 0.0486 |
| 0.049 | 248.62 | 45000 | 0.0486 |
| 0.0488 | 276.24 | 50000 | 0.0486 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.13.1+cu117
- Datasets 2.8.0
- Tokenizers 0.13.2