roberta-fake-news-detector-2

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

• Loss: 0.2056
• Accuracy: 0.9439
• Precision: 0.9444
• Recall: 0.9473
• F1: 0.9459
• Confusion Matrix: [[18282, 1160], [1097, 19721]]

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: 4e-05
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 128
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1	Confusion Matrix
0.2294	1.0	661	0.2232	0.9331	0.9239	0.9483	0.9360	[[4176, 378], [250, 4590]]
0.2292	2.0	1322	0.2074	0.9415	0.9339	0.9539	0.9438	[[4227, 327], [223, 4617]]
0.1617	3.0	1983	0.2132	0.9437	0.9478	0.9426	0.9452	[[4303, 251], [278, 4562]]
0.1948	4.0	2644	0.2052	0.9439	0.9435	0.9479	0.9457	[[4279, 275], [252, 4588]]
0.1469	5.0	3305	0.2212	0.9418	0.9490	0.9374	0.9431	[[4310, 244], [303, 4537]]
0.1413	6.0	3966	0.2179	0.9412	0.9457	0.9399	0.9428	[[4293, 261], [291, 4549]]
0.162	7.0	4627	0.2229	0.9437	0.9431	0.9479	0.9455	[[4277, 277], [252, 4588]]

Framework versions

• Transformers 4.51.3
• Pytorch 2.5.1
• Datasets 3.5.0
• Tokenizers 0.21.1