alt-gnome
/

spam-detector-v0.2

Text Classification

Model card Files Files and versions

🚀 This model is a fine-tuned version of RUSpam/spam_deberta_v4 on a custom dataset of Russian-language messages.

Training configuration:

Optimizer: AdamW
Learning rate: 2e-5
Batch size: 16
Epochs: 3
Metrics tracked: accuracy

License

This model is released under the MIT license.
It is based on RUSpam/spam_deberta_v4, which is also licensed under MIT.

Downloads last month: 293

Safetensors

Model size

82.1M params

Tensor type

F32

·

Model tree for alt-gnome/spam-detector-v0.2

Base model

RUSpam/spam_deberta_v4

Finetuned

(2)

this model

Dataset used to train alt-gnome/spam-detector-v0.2