Roberta_fakecovid

This model is a fine-tuned version of FacebookAI/xlm-roberta-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0417
• Accuracy: 0.988
• Auc: 0.941
• Precision: 0.0
• Recall: 0.0
• F1: 0.0
• F1-macro: 0.497
• F1-micro: 0.988
• F1-weighted: 0.982

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: 2e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 2
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Auc	Precision	Recall	F1	F1-macro	F1-micro	F1-weighted
0.1257	0.3017	50	0.0617	0.988	0.913	0.0	0.0	0.0	0.497	0.988	0.982
0.0482	0.6033	100	0.0696	0.988	0.944	0.0	0.0	0.0	0.497	0.988	0.982
0.0557	0.9050	150	0.0717	0.988	0.892	0.0	0.0	0.0	0.497	0.988	0.982
0.0419	1.2051	200	0.0547	0.988	0.909	0.0	0.0	0.0	0.497	0.988	0.982
0.051	1.5068	250	0.0472	0.988	0.916	0.0	0.0	0.0	0.497	0.988	0.982
0.0571	1.8084	300	0.0417	0.988	0.941	0.0	0.0	0.0	0.497	0.988	0.982

Framework versions

• Transformers 4.55.2
• Pytorch 2.8.0+cu126
• Datasets 4.0.0
• Tokenizers 0.21.4