metadata
license: mit
tags:
- generated_from_trainer
model-index:
- name: gpt2-kl_001_07_hscnspecial-hs_cn
results: []
gpt2-kl_001_07_hscnspecial-hs_cn
This model is a fine-tuned version of gpt2-medium on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.5583
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 |
|---|---|---|---|
| 73.5264 | 0.02 | 10 | 69.5797 |
| 46.0831 | 0.04 | 20 | 32.9588 |
| 13.5437 | 0.06 | 30 | 10.6394 |
| 6.7905 | 0.08 | 40 | 4.1876 |
| 3.5385 | 0.1 | 50 | 2.0711 |
| 1.5427 | 0.12 | 60 | 1.1384 |
| 1.1141 | 0.14 | 70 | 0.8584 |
| 0.926 | 0.16 | 80 | 0.7673 |
| 0.7432 | 0.18 | 90 | 0.6549 |
| 0.7195 | 0.2 | 100 | 0.6203 |
| 0.6079 | 0.22 | 110 | 0.5966 |
| 0.6356 | 0.24 | 120 | 0.5855 |
| 0.6443 | 0.26 | 130 | 0.5837 |
| 0.6171 | 0.28 | 140 | 0.5791 |
| 0.6031 | 0.3 | 150 | 0.5869 |
| 0.587 | 0.32 | 160 | 0.5674 |
| 0.518 | 0.34 | 170 | 0.5665 |
| 0.6313 | 0.36 | 180 | 0.5698 |
| 0.6331 | 0.38 | 190 | 0.5624 |
| 0.5768 | 0.4 | 200 | 0.5575 |
| 0.6768 | 0.42 | 210 | 0.5587 |
| 0.6288 | 0.44 | 220 | 0.5500 |
| 0.5371 | 0.46 | 230 | 0.5510 |
| 0.5499 | 0.48 | 240 | 0.5578 |
| 0.548 | 0.5 | 250 | 0.5583 |
Framework versions
- Transformers 4.29.0.dev0
- Pytorch 1.12.0a0+bd13bc6
- Datasets 2.12.0
- Tokenizers 0.13.3