README.md

metadata

library_name: transformers
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - f1
  - precision
  - recall
model-index:
  - name: hate_speech
    results: []

hate_speech

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.6154
• Accuracy: 0.7362
• Auc Score: 0.8016
• F1: 0.7714
• Precision: 0.7557
• Recall: 0.7878

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: 3e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Use OptimizerNames.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: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Auc Score	F1	Precision	Recall
0.6869	0.1845	100	0.6557	0.6338	0.6812	0.7327	0.6235	0.8882
0.655	0.3690	200	0.6146	0.6527	0.7396	0.6440	0.7652	0.5559
0.6269	0.5535	300	0.5991	0.6850	0.7517	0.7065	0.7459	0.6710
0.6089	0.7380	400	0.6444	0.6485	0.7632	0.6182	0.8003	0.5037
0.5962	0.9225	500	0.6072	0.6827	0.7764	0.7673	0.6551	0.9257
0.5633	1.1070	600	0.5957	0.6960	0.7767	0.7685	0.6745	0.8931
0.5067	1.2915	700	0.6133	0.7196	0.7900	0.7778	0.7042	0.8686
0.505	1.4760	800	0.5766	0.7076	0.7703	0.7414	0.7408	0.7420
0.5115	1.6605	900	0.5415	0.7348	0.8011	0.7823	0.7295	0.8433
0.5027	1.8450	1000	0.5837	0.7163	0.7949	0.7755	0.7015	0.8669
0.4695	2.0295	1100	0.5699	0.7173	0.7889	0.7686	0.7149	0.8310
0.4032	2.2140	1200	0.6357	0.7145	0.7865	0.7718	0.7036	0.8547
0.3833	2.3985	1300	0.6236	0.7292	0.8014	0.7697	0.7409	0.8008
0.3924	2.5830	1400	0.6533	0.7103	0.7970	0.7784	0.6855	0.9004
0.3673	2.7675	1500	0.6303	0.7260	0.8004	0.7729	0.7268	0.8253
0.3749	2.9520	1600	0.6154	0.7362	0.8016	0.7714	0.7557	0.7878

Framework versions

• Transformers 4.52.4
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.1