<!-- 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. -->
small-mlm-glue-wnli
This model is a fine-tuned version of google/bert_uncased_L-4_H-512_A-8 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1284
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: 3e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- num_epochs: 200
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
1.7452 | 6.25 | 500 | 1.2770 |
0.9127 | 12.5 | 1000 | 0.8006 |
0.6024 | 18.75 | 1500 | 0.5714 |
0.3967 | 25.0 | 2000 | 0.6533 |
0.3443 | 31.25 | 2500 | 0.3623 |
0.2739 | 37.5 | 3000 | 0.3035 |
0.2326 | 43.75 | 3500 | 0.2767 |
0.1942 | 50.0 | 4000 | 0.1730 |
0.1666 | 56.25 | 4500 | 0.1674 |
0.1688 | 62.5 | 5000 | 0.1459 |
0.1378 | 68.75 | 5500 | 0.2353 |
0.1344 | 75.0 | 6000 | 0.1074 |
0.1259 | 81.25 | 6500 | 0.1757 |
0.1176 | 87.5 | 7000 | 0.0720 |
0.1114 | 93.75 | 7500 | 0.1377 |
0.0993 | 100.0 | 8000 | 0.1752 |
0.0992 | 106.25 | 8500 | 0.1284 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu116
- Datasets 2.8.1.dev0
- Tokenizers 0.13.2