<!-- 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-ear_001-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: 0.5463
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 |
---|---|---|---|
73.4859 | 0.02 | 10 | 69.5673 |
46.0211 | 0.04 | 20 | 32.9286 |
13.5057 | 0.06 | 30 | 10.6409 |
6.7449 | 0.08 | 40 | 4.2311 |
2.6783 | 0.1 | 50 | 2.0356 |
1.4651 | 0.12 | 60 | 1.0993 |
1.0415 | 0.14 | 70 | 0.9246 |
0.856 | 0.16 | 80 | 0.7304 |
0.7255 | 0.18 | 90 | 0.6521 |
0.7163 | 0.2 | 100 | 0.7056 |
0.5533 | 0.22 | 110 | 0.6339 |
0.6175 | 0.24 | 120 | 0.5969 |
0.5876 | 0.26 | 130 | 0.5812 |
0.5815 | 0.28 | 140 | 0.5724 |
0.5549 | 0.3 | 150 | 0.5688 |
0.5499 | 0.32 | 160 | 0.5735 |
0.499 | 0.34 | 170 | 0.5637 |
0.6024 | 0.36 | 180 | 0.5580 |
0.5936 | 0.38 | 190 | 0.5556 |
0.5448 | 0.4 | 200 | 0.5528 |
0.6419 | 0.42 | 210 | 0.5535 |
0.5966 | 0.44 | 220 | 0.5508 |
0.5029 | 0.46 | 230 | 0.5513 |
0.5079 | 0.48 | 240 | 0.5518 |
0.5117 | 0.5 | 250 | 0.5498 |
0.5132 | 0.52 | 260 | 0.5517 |
0.518 | 0.54 | 270 | 0.5435 |
0.5532 | 0.56 | 280 | 0.5439 |
0.5425 | 0.58 | 290 | 0.5450 |
0.5104 | 0.6 | 300 | 0.5463 |
Framework versions
- Transformers 4.29.0.dev0
- Pytorch 1.12.0a0+bd13bc6
- Datasets 2.12.0
- Tokenizers 0.13.3