afaji/summerschool-bert-hate

summerschool-1layer-distill-irony

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: 1.6811
• Accuracy: 0.5226

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: 0.0001
• 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: 2

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.5919	0.1776	100	0.5602	0.718
0.5094	0.3552	200	0.5458	0.722
0.458	0.5329	300	0.5431	0.758
0.4826	0.7105	400	0.5628	0.758
0.4127	0.8881	500	0.5009	0.751
0.3931	1.0657	600	0.5868	0.751
0.287	1.2433	700	0.6045	0.767
0.262	1.4210	800	0.5380	0.771
0.2592	1.5986	900	0.5837	0.779
0.2711	1.7762	1000	0.5107	0.782
0.276	1.9538	1100	0.4911	0.796

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1