7c249ee43c430aa74ce9eb72b8c9807f

This model is a fine-tuned version of albert/albert-xxlarge-v2 on the contemmcm/hate-speech-and-offensive-language dataset. It achieves the following results on the evaluation set:

• Loss: 0.6763
• Data Size: 1.0
• Epoch Runtime: 56.8902
• 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	1.2309	0	4.1677	0.3452	0.2711
No log	1	619	0.7273	0.0078	4.6957	0.7364	0.3367
No log	2	1238	0.6758	0.0156	5.0997	0.7668	0.2917
0.0174	3	1857	0.6966	0.0312	6.1633	0.7672	0.2894
0.0174	4	2476	0.6967	0.0625	7.8329	0.7672	0.2894
0.6886	5	3095	0.6717	0.125	11.1219	0.7672	0.2894
0.0673	6	3714	0.6901	0.25	17.6613	0.7672	0.2894
0.6886	7	4333	0.6802	0.5	30.7308	0.7672	0.2894
0.7114	8.0	4952	0.6904	1.0	57.1601	0.7672	0.2894
0.6769	9.0	5571	0.6763	1.0	56.8902	0.7672	0.2894

Framework versions

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