distilbert-synth-vishing

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

• Loss: 0.0043
• Accuracy: 0.9989
• F1: 0.9988
• Precision: 1.0
• Recall: 0.9976

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: 32
• eval_batch_size: 16
• seed: 42
• 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: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
0.2075	0.2286	80	0.0510	0.9853	0.9849	0.9702	1.0
0.0347	0.4571	160	0.0160	0.9959	0.9957	0.9922	0.9992
0.0196	0.6857	240	0.0122	0.9962	0.9961	0.9953	0.9968
0.0143	0.9143	320	0.0097	0.9970	0.9968	0.9976	0.9961
0.0112	1.1429	400	0.0054	0.9985	0.9984	0.9969	1.0
0.0101	1.3714	480	0.0034	0.9989	0.9988	0.9976	1.0
0.0032	1.6	560	0.0043	0.9989	0.9988	1.0	0.9976
0.0030	1.8286	640	0.0036	0.9992	0.9992	0.9992	0.9992
0.0047	2.0571	720	0.0022	0.9992	0.9992	0.9984	1.0
0.0036	2.2857	800	0.0037	0.9989	0.9988	0.9992	0.9984
0.0005	2.5143	880	0.0043	0.9989	0.9988	1.0	0.9976

Framework versions

• Transformers 5.0.0
• Pytorch 2.11.0+cu128
• Datasets 4.8.5
• Tokenizers 0.22.2