<!-- 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_04-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: 0.5395
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 |
---|---|---|---|
73.8253 | 0.02 | 10 | 69.5735 |
46.3767 | 0.04 | 20 | 33.0039 |
13.8102 | 0.06 | 30 | 10.6426 |
7.0975 | 0.08 | 40 | 4.2638 |
3.8614 | 0.1 | 50 | 2.0637 |
1.707 | 0.12 | 60 | 1.1069 |
1.3278 | 0.14 | 70 | 0.8700 |
1.1456 | 0.16 | 80 | 0.7454 |
1.0354 | 0.18 | 90 | 0.6439 |
0.9738 | 0.2 | 100 | 0.6463 |
0.8737 | 0.22 | 110 | 0.6093 |
0.8784 | 0.24 | 120 | 0.5951 |
0.8958 | 0.26 | 130 | 0.6124 |
0.9413 | 0.28 | 140 | 0.5783 |
0.9534 | 0.3 | 150 | 0.5726 |
0.893 | 0.32 | 160 | 0.5745 |
0.7813 | 0.34 | 170 | 0.5689 |
0.9262 | 0.36 | 180 | 0.5700 |
0.9613 | 0.38 | 190 | 0.5610 |
0.83 | 0.4 | 200 | 0.5601 |
0.9942 | 0.42 | 210 | 0.5601 |
0.9548 | 0.44 | 220 | 0.5549 |
0.7824 | 0.46 | 230 | 0.5575 |
0.8444 | 0.48 | 240 | 0.5560 |
0.8078 | 0.5 | 250 | 0.5519 |
0.8268 | 0.52 | 260 | 0.5490 |
0.8238 | 0.54 | 270 | 0.5477 |
0.8885 | 0.56 | 280 | 0.5500 |
0.8366 | 0.58 | 290 | 0.5471 |
0.7629 | 0.6 | 300 | 0.5484 |
0.8435 | 0.62 | 310 | 0.5463 |
0.9687 | 0.64 | 320 | 0.5458 |
0.7741 | 0.66 | 330 | 0.5472 |
0.7782 | 0.68 | 340 | 0.5446 |
0.7781 | 0.7 | 350 | 0.5521 |
0.7885 | 0.72 | 360 | 0.5511 |
0.8956 | 0.74 | 370 | 0.5389 |
0.8358 | 0.76 | 380 | 0.5379 |
0.8112 | 0.78 | 390 | 0.5346 |
0.8803 | 0.8 | 400 | 0.5337 |
0.7743 | 0.82 | 410 | 0.5371 |
0.8069 | 0.84 | 420 | 0.5388 |
0.8662 | 0.86 | 430 | 0.5332 |
0.8161 | 0.88 | 440 | 0.5354 |
0.8506 | 0.9 | 450 | 0.5319 |
0.7697 | 0.92 | 460 | 0.5314 |
0.7628 | 0.94 | 470 | 0.5309 |
0.835 | 0.96 | 480 | 0.5304 |
0.8235 | 0.98 | 490 | 0.5303 |
0.7438 | 1.0 | 500 | 0.5292 |
0.6762 | 1.02 | 510 | 0.5377 |
0.7267 | 1.04 | 520 | 0.5381 |
0.6736 | 1.06 | 530 | 0.5395 |
Framework versions
- Transformers 4.28.0
- Pytorch 1.11.0+cu113
- Datasets 2.11.0
- Tokenizers 0.12.1