bert_uncased_L-4_H-128_A-2_wnli

This model is a fine-tuned version of google/bert_uncased_L-4_H-128_A-2 on the GLUE WNLI dataset. It achieves the following results on the evaluation set:

• Loss: 0.7042
• Accuracy: 0.3521

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: 256
• eval_batch_size: 256
• seed: 10
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6968	1.0	3	0.7060	0.4225
0.6898	2.0	6	0.7042	0.3521
0.6912	3.0	9	0.7055	0.3662
0.695	4.0	12	0.7075	0.3380
0.6946	5.0	15	0.7099	0.3099
0.6921	6.0	18	0.7122	0.3099
0.6874	7.0	21	0.7140	0.3239

Framework versions

• Transformers 4.46.3
• Pytorch 2.2.1+cu118
• Datasets 2.17.0
• Tokenizers 0.20.3