metadata
license: apache-2.0
tags:
- fairsteer
- bias-detection
- tinyllama
- safetensors
library_name: safetensors
pipeline_tag: text-classification
BAD Classifier for FairSteer
Biased Activation Detection (BAD) classifier for TinyLlama-1.1B.
Artifacts
- Model:
model.safetensors(SafeTensors format) - Scaler:
scaler.pkl(StandardScaler) - Config:
config.json
Stats
- Balanced Accuracy: 73.58%
- Best Layer: 17
- Training Date: 2025-12-12