bert-finetuned-20newsgroups

This model is a fine-tuned version of bert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.9917

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: 4
• eval_batch_size: 8
• seed: 42
• 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
• lr_scheduler_warmup_steps: 10
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss
1.5307	0.2762	600	1.1779
1.0973	0.5525	1200	0.9778
1.0372	0.8287	1800	0.9016
0.8659	1.1050	2400	0.9472
0.6981	1.3812	3000	0.9557
0.6827	1.6575	3600	1.0130
0.6297	1.9337	4200	0.9283
0.4627	2.2099	4800	0.9689
0.3674	2.4862	5400	1.0597
0.4024	2.7624	6000	0.9917

Framework versions

• Transformers 4.46.3
• Pytorch 2.5.1+cu121
• Tokenizers 0.20.3