8c9c7da340e2d29412f6d4d2edf61526

This model is a fine-tuned version of FacebookAI/xlm-roberta-large-finetuned-conll02-spanish on the contemmcm/hate-speech-and-offensive-language dataset. It achieves the following results on the evaluation set:

• Loss: 0.6807
• Data Size: 1.0
• Epoch Runtime: 106.9286
• 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	0.8513	0	7.5714	0.7666	0.2901
No log	1	619	0.7465	0.0078	8.5869	0.7672	0.2894
No log	2	1238	0.6772	0.0156	10.2448	0.7672	0.2894
0.016	3	1857	0.7231	0.0312	12.5154	0.8433	0.5092
0.016	4	2476	0.4619	0.0625	15.7105	0.8807	0.5812
0.4457	5	3095	0.3738	0.125	22.7781	0.8914	0.5927
0.0506	6	3714	0.5484	0.25	35.3805	0.7672	0.2894
0.4147	7	4333	0.3867	0.5	60.3998	0.8860	0.5871
0.6808	8.0	4952	0.6804	1.0	105.9201	0.7672	0.2894
0.6484	9.0	5571	0.6807	1.0	106.9286	0.7672	0.2894

Framework versions

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