bert-base-uncased-finetuned-swag

This model is a fine-tuned version of bert-base-uncased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.1649
• Accuracy: 0.4857

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	26	1.3860	0.3238
No log	2.0	52	1.3833	0.3238
No log	3.0	78	1.3547	0.3810
No log	4.0	104	1.3130	0.3905
No log	5.0	130	1.3733	0.4000
No log	6.0	156	1.6432	0.4476
No log	7.0	182	1.8118	0.4952
No log	8.0	208	2.0408	0.4571
No log	9.0	234	1.9043	0.4952
No log	10.0	260	1.9755	0.4952
No log	11.0	286	2.0813	0.4857
No log	12.0	312	2.0578	0.4571
No log	13.0	338	2.0979	0.4762
No log	14.0	364	2.1136	0.4857
No log	15.0	390	2.1649	0.4857

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.15.0
• Tokenizers 0.15.0