5b158cd788c0f5b86aa5bd2a08320559

This model is a fine-tuned version of studio-ousia/mluke-large on the contemmcm/hate-speech-and-offensive-language dataset. It achieves the following results on the evaluation set:

• Loss: 0.6767
• Data Size: 1.0
• Epoch Runtime: 113.0819
• Accuracy: 0.7672
• F1 Macro: 0.2894

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: 8
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• total_train_batch_size: 32
• total_eval_batch_size: 32
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: constant
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Data Size	Epoch Runtime	Accuracy	F1 Macro
No log	0	0	1.0790	0	8.4248	0.4706	0.3015
No log	1	619	0.7282	0.0078	9.4745	0.7672	0.2894
No log	2	1238	0.6338	0.0156	11.0407	0.7672	0.2894
0.0155	3	1857	0.4590	0.0312	13.7676	0.7672	0.2894
0.0155	4	2476	0.6710	0.0625	17.6240	0.7672	0.2894
0.5533	5	3095	0.4396	0.125	25.1486	0.8444	0.5431
0.0472	6	3714	0.4999	0.25	37.8147	0.8285	0.5229
0.4375	7	4333	0.4445	0.5	63.8339	0.8214	0.5260
0.571	8.0	4952	0.6792	1.0	117.8540	0.7672	0.2894
0.6404	9.0	5571	0.6767	1.0	113.0819	0.7672	0.2894

Framework versions

• Transformers 4.57.0
• Pytorch 2.8.0+cu128
• Datasets 4.3.0
• Tokenizers 0.22.1