<!-- 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. -->
add_BERT_48_wnli
This model is a fine-tuned version of gokuls/add_bert_12_layer_model_complete_training_new_48 on the GLUE WNLI dataset. It achieves the following results on the evaluation set:
- Loss: 0.6779
- Accuracy: 0.5634
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: 4e-05
- train_batch_size: 128
- eval_batch_size: 128
- seed: 10
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 50
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
0.8653 | 1.0 | 5 | 0.7214 | 0.4366 |
0.7161 | 2.0 | 10 | 0.7479 | 0.4366 |
0.7092 | 3.0 | 15 | 0.6853 | 0.5352 |
0.7061 | 4.0 | 20 | 0.6955 | 0.4507 |
0.7136 | 5.0 | 25 | 0.6992 | 0.5634 |
0.7174 | 6.0 | 30 | 0.6863 | 0.5634 |
0.708 | 7.0 | 35 | 0.7285 | 0.4366 |
0.7179 | 8.0 | 40 | 0.6839 | 0.5634 |
0.7647 | 9.0 | 45 | 0.6923 | 0.5634 |
0.7065 | 10.0 | 50 | 0.7315 | 0.4648 |
0.7026 | 11.0 | 55 | 0.6779 | 0.5634 |
0.6973 | 12.0 | 60 | 0.6849 | 0.5634 |
0.6934 | 13.0 | 65 | 0.6934 | 0.4789 |
0.6978 | 14.0 | 70 | 0.7005 | 0.4366 |
0.6906 | 15.0 | 75 | 0.6849 | 0.5634 |
0.6927 | 16.0 | 80 | 0.6959 | 0.4648 |
Framework versions
- Transformers 4.30.2
- Pytorch 1.14.0a0+410ce96
- Datasets 2.13.0
- Tokenizers 0.13.3