distilbert_sa_GLUE_Experiment_wnli_384

This model is a fine-tuned version of distilbert-base-uncased on the GLUE WNLI dataset. It achieves the following results on the evaluation set:

• Loss: 0.6861
• 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: 5e-05
• train_batch_size: 256
• eval_batch_size: 256
• 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
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.7012	1.0	3	0.7047	0.4366
0.7043	2.0	6	0.6952	0.4366
0.6928	3.0	9	0.6861	0.5634
0.6998	4.0	12	0.6874	0.5634
0.6927	5.0	15	0.6957	0.4366
0.6952	6.0	18	0.7001	0.4366
0.6966	7.0	21	0.6927	0.5634
0.6917	8.0	24	0.6909	0.5634

Framework versions

• Transformers 4.26.0
• Pytorch 1.14.0a0+410ce96
• Datasets 2.8.0
• Tokenizers 0.13.2