599959e23c0ba7a09e34eceb2caf003e

This model is a fine-tuned version of albert/albert-xxlarge-v1 on the contemmcm/hate-speech-and-offensive-language dataset. It achieves the following results on the evaluation set:

• Loss: 0.6817
• Data Size: 0.25
• Epoch Runtime: 14.4440
• 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	0.9929	0	3.2500	0.6749	0.3267
No log	1	619	0.6891	0.0078	3.6597	0.7650	0.2942
No log	2	1238	0.6747	0.0156	4.0273	0.7672	0.2894
0.0165	3	1857	0.7269	0.0312	4.9218	0.7672	0.2894
0.0165	4	2476	0.6903	0.0625	6.1881	0.7672	0.2894
0.6723	5	3095	0.6878	0.125	8.9158	0.7672	0.2894
0.0635	6	3714	0.6817	0.25	14.4440	0.7672	0.2894

Framework versions

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