Fetching the paper…

Bias A-head? Analyzing Bias in Transformer-Based Language Model Attention Heads · Around