fine_tuned_spam_model

This model is a fine-tuned version of huawei-noah/TinyBERT_General_4L_312D on a dataset of batch-labeled emails and SMS messages that were identified to be spam (Enron, spamassassin, sms-spam, etc.). It achieves the following results on the evaluation set:

• Loss: 0.6292
• Accuracy: 0.7664

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• 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: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6875	1.0	364	0.7508	0.7226
0.5574	2.0	728	0.6804	0.7292
0.5481	3.0	1092	0.6292	0.7664

Framework versions

• Transformers 4.49.0
• Pytorch 2.6.0+cpu
• Datasets 3.3.2
• Tokenizers 0.21.1