all-base-len
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 4.8452
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: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 1000
- num_epochs: 6
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 6.4465 | 0.31 | 500 | 5.5455 |
| 5.1194 | 0.62 | 1000 | 5.1702 |
| 4.7736 | 0.94 | 1500 | 5.0228 |
| 4.5064 | 1.25 | 2000 | 4.9752 |
| 4.3839 | 1.56 | 2500 | 4.8636 |
| 4.2864 | 1.87 | 3000 | 4.7875 |
| 4.1137 | 2.19 | 3500 | 4.7734 |
| 4.0249 | 2.5 | 4000 | 4.7316 |
| 3.9841 | 2.81 | 4500 | 4.7210 |
| 3.8607 | 3.12 | 5000 | 4.7148 |
| 3.7214 | 3.44 | 5500 | 4.7177 |
| 3.7078 | 3.75 | 6000 | 4.6981 |
| 3.6456 | 4.06 | 6500 | 4.7174 |
| 3.4475 | 4.37 | 7000 | 4.7426 |
| 3.4411 | 4.68 | 7500 | 4.7503 |
| 3.4265 | 5.0 | 8000 | 4.7376 |
| 3.2581 | 5.31 | 8500 | 4.7947 |
| 3.2567 | 5.62 | 9000 | 4.7966 |
| 3.2505 | 5.93 | 9500 | 4.7984 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
- Downloads last month
- 5
docker model run hf.co/NasimB/all-base-len