<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
gpt2-kl_1_06-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: 0.5352
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 |
|---|---|---|---|
| 74.175 | 0.02 | 10 | 69.5742 |
| 46.6833 | 0.04 | 20 | 32.9573 |
| 14.147 | 0.06 | 30 | 10.6344 |
| 7.4081 | 0.08 | 40 | 4.2202 |
| 4.1981 | 0.1 | 50 | 2.0325 |
| 1.9903 | 0.12 | 60 | 1.0829 |
| 1.5743 | 0.14 | 70 | 0.8595 |
| 1.4537 | 0.16 | 80 | 0.7859 |
| 1.4022 | 0.18 | 90 | 0.7212 |
| 1.3332 | 0.2 | 100 | 0.6727 |
| 1.1681 | 0.22 | 110 | 0.6850 |
| 1.1991 | 0.24 | 120 | 0.5978 |
| 1.2297 | 0.26 | 130 | 0.6137 |
| 1.3475 | 0.28 | 140 | 0.5888 |
| 1.3468 | 0.3 | 150 | 0.5783 |
| 1.2516 | 0.32 | 160 | 0.5765 |
| 1.1055 | 0.34 | 170 | 0.5752 |
| 1.2874 | 0.36 | 180 | 0.5684 |
| 1.3511 | 0.38 | 190 | 0.5613 |
| 1.1492 | 0.4 | 200 | 0.5571 |
| 1.3802 | 0.42 | 210 | 0.5567 |
| 1.3072 | 0.44 | 220 | 0.5527 |
| 1.1026 | 0.46 | 230 | 0.5538 |
| 1.199 | 0.48 | 240 | 0.5497 |
| 1.1124 | 0.5 | 250 | 0.5513 |
| 1.1861 | 0.52 | 260 | 0.5495 |
| 1.1603 | 0.54 | 270 | 0.5434 |
| 1.2407 | 0.56 | 280 | 0.5451 |
| 1.1338 | 0.58 | 290 | 0.5437 |
| 1.0556 | 0.6 | 300 | 0.5428 |
| 1.2218 | 0.62 | 310 | 0.5392 |
| 1.3505 | 0.64 | 320 | 0.5408 |
| 1.1001 | 0.66 | 330 | 0.5426 |
| 1.1123 | 0.68 | 340 | 0.5385 |
| 1.1046 | 0.7 | 350 | 0.5385 |
| 1.1291 | 0.72 | 360 | 0.5383 |
| 1.2087 | 0.74 | 370 | 0.5378 |
| 1.1888 | 0.76 | 380 | 0.5380 |
| 1.1634 | 0.78 | 390 | 0.5363 |
| 1.249 | 0.8 | 400 | 0.5351 |
| 1.1197 | 0.82 | 410 | 0.5350 |
| 1.1508 | 0.84 | 420 | 0.5366 |
| 1.2025 | 0.86 | 430 | 0.5340 |
| 1.13 | 0.88 | 440 | 0.5347 |
| 1.1664 | 0.9 | 450 | 0.5371 |
| 1.048 | 0.92 | 460 | 0.5352 |
Framework versions
- Transformers 4.28.0
- Pytorch 1.11.0+cu113
- Datasets 2.11.0
- Tokenizers 0.12.1
