noise_test_v1

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: 0.2388
• Accuracy Offensive: 0.9434
• F1 Offensive: 0.9423
• Accuracy Targeted: 0.9350
• F1 Targeted: 0.9128
• Accuracy Stance: 0.8776
• F1 Stance: 0.8351

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: 1e-05
• train_batch_size: 8
• eval_batch_size: 8
• 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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy Offensive	F1 Offensive	Accuracy Targeted	F1 Targeted	Accuracy Stance	F1 Stance
0.7254	1.0	1324	0.6149	0.6556	0.5204	0.6941	0.6005	0.7009	0.5777
0.4464	2.0	2648	0.3609	0.8376	0.8216	0.9290	0.9068	0.8727	0.8286
0.3741	3.0	3972	0.3068	0.8935	0.8878	0.9324	0.9102	0.8735	0.8308
0.3488	4.0	5296	0.2872	0.9124	0.9088	0.9332	0.9109	0.8761	0.8326
0.31	5.0	6620	0.2674	0.9362	0.9346	0.9369	0.9146	0.875	0.8337
0.3162	6.0	7944	0.2723	0.9320	0.9302	0.9354	0.9131	0.8758	0.8331
0.3172	7.0	9268	0.2606	0.9324	0.9306	0.9320	0.9098	0.8773	0.8339
0.2963	8.0	10592	0.2406	0.9452	0.9441	0.9392	0.9168	0.8784	0.8360
0.2856	9.0	11916	0.2465	0.9403	0.9390	0.9350	0.9128	0.8776	0.8348
0.2749	10.0	13240	0.2369	0.9460	0.9450	0.9373	0.9150	0.8784	0.8361
0.2736	11.0	14564	0.2419	0.9505	0.9497	0.9384	0.9161	0.8712	0.8304
0.2753	12.0	15888	0.2513	0.9494	0.9487	0.9347	0.9124	0.8663	0.8262

Framework versions

• Transformers 4.50.2
• Pytorch 2.6.0+cu124
• Datasets 3.0.1
• Tokenizers 0.21.1