<!-- 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_05-hs_cn_decay
This model is a fine-tuned version of gpt2-medium on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.5428
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 |
---|---|---|---|
73.9843 | 0.02 | 10 | 69.5786 |
46.5197 | 0.04 | 20 | 32.9833 |
13.9523 | 0.06 | 30 | 10.6378 |
7.2147 | 0.08 | 40 | 4.2578 |
3.4244 | 0.1 | 50 | 2.0277 |
1.8713 | 0.12 | 60 | 1.0949 |
1.5036 | 0.14 | 70 | 0.9420 |
1.2167 | 0.16 | 80 | 0.7558 |
1.1629 | 0.18 | 90 | 0.6537 |
1.0997 | 0.2 | 100 | 0.6998 |
1.0395 | 0.22 | 110 | 0.5946 |
0.9896 | 0.24 | 120 | 0.5825 |
1.0067 | 0.26 | 130 | 0.5759 |
1.0529 | 0.28 | 140 | 0.5738 |
1.072 | 0.3 | 150 | 0.5701 |
0.9853 | 0.32 | 160 | 0.5668 |
0.8855 | 0.34 | 170 | 0.5685 |
1.0401 | 0.36 | 180 | 0.5600 |
1.0669 | 0.38 | 190 | 0.5558 |
0.9189 | 0.4 | 200 | 0.5505 |
1.1044 | 0.42 | 210 | 0.5526 |
1.0624 | 0.44 | 220 | 0.5470 |
0.8628 | 0.46 | 230 | 0.5493 |
0.9359 | 0.48 | 240 | 0.5452 |
0.8814 | 0.5 | 250 | 0.5466 |
0.9381 | 0.52 | 260 | 0.5447 |
0.8973 | 0.54 | 270 | 0.5418 |
0.9575 | 0.56 | 280 | 0.5438 |
0.8965 | 0.58 | 290 | 0.5434 |
0.837 | 0.6 | 300 | 0.5428 |
Framework versions
- Transformers 4.28.0
- Pytorch 1.11.0+cu113
- Datasets 2.11.0
- Tokenizers 0.12.1