gpt2-synth-real
This model is a fine-tuned version of gpt2-medium on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.4267
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: 4
- seed: 21
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 3.0
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 80.0811 | 0.01 | 10 | 76.0005 |
| 63.2867 | 0.02 | 20 | 55.5177 |
| 30.7747 | 0.03 | 30 | 19.9807 |
| 17.8136 | 0.04 | 40 | 12.4432 |
| 6.5809 | 0.05 | 50 | 5.7868 |
| 3.8801 | 0.06 | 60 | 3.3238 |
| 2.2807 | 0.07 | 70 | 1.8328 |
| 1.3763 | 0.08 | 80 | 1.1114 |
| 0.8658 | 0.09 | 90 | 1.0464 |
| 0.748 | 0.1 | 100 | 0.8199 |
| 0.5694 | 0.12 | 110 | 0.5984 |
| 0.6427 | 0.13 | 120 | 0.6168 |
| 0.5534 | 0.14 | 130 | 0.5981 |
| 0.5483 | 0.15 | 140 | 0.5450 |
| 0.5384 | 0.16 | 150 | 0.4926 |
| 0.4926 | 0.17 | 160 | 0.4924 |
| 0.5059 | 0.18 | 170 | 0.4762 |
| 0.4372 | 0.19 | 180 | 0.4733 |
| 0.4833 | 0.2 | 190 | 0.4785 |
| 0.4511 | 0.21 | 200 | 0.4511 |
| 0.4163 | 0.22 | 210 | 0.4534 |
| 0.4849 | 0.23 | 220 | 0.4420 |
| 0.4857 | 0.24 | 230 | 0.4421 |
| 0.6163 | 0.25 | 240 | 0.4336 |
| 0.5151 | 0.26 | 250 | 0.4344 |
| 0.4533 | 0.27 | 260 | 0.4280 |
| 0.3812 | 0.28 | 270 | 0.4387 |
| 0.475 | 0.29 | 280 | 0.4274 |
| 0.4354 | 0.3 | 290 | 0.4272 |
| 0.3912 | 0.31 | 300 | 0.4305 |
| 0.3944 | 0.32 | 310 | 0.4204 |
| 0.4166 | 0.33 | 320 | 0.4239 |
| 0.3677 | 0.35 | 330 | 0.4246 |
| 0.4135 | 0.36 | 340 | 0.4267 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.11.0+cu113
- Datasets 2.6.1
- Tokenizers 0.12.1