<!-- 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. -->
tiny-mlm-conll2003
This model is a fine-tuned version of google/bert_uncased_L-2_H-128_A-2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 4.6472
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 |
---|---|---|---|
5.4619 | 0.4 | 500 | 5.1363 |
5.2804 | 0.8 | 1000 | 4.9579 |
5.1847 | 1.2 | 1500 | 4.7870 |
5.0376 | 1.6 | 2000 | 4.8272 |
4.9622 | 2.0 | 2500 | 4.8460 |
4.913 | 2.4 | 3000 | 4.7859 |
4.9216 | 2.8 | 3500 | 4.7816 |
4.8176 | 3.2 | 4000 | 4.7040 |
4.7735 | 3.6 | 4500 | 4.6986 |
4.7892 | 4.0 | 5000 | 4.6846 |
4.718 | 4.4 | 5500 | 4.6943 |
4.6564 | 4.8 | 6000 | 4.6461 |
4.6593 | 5.2 | 6500 | 4.6011 |
4.6384 | 5.6 | 7000 | 4.7286 |
4.5704 | 6.0 | 7500 | 4.7027 |
4.478 | 6.4 | 8000 | 4.6472 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu116
- Datasets 2.8.1.dev0
- Tokenizers 0.13.2