This is a repository with the code for the ACL 2019 paper "Analyzing Multi-Head Self-Attention: Specialized Heads Do the Heavy Lifting, the Rest Can Be Pruned" and the ACL 2021 paper "Analyzing Source and Target Contributions to NMT Predictions".
CLI toolkit that ingests qk-sniffer dumps, measures per-head positional predictability and attention plasticity, and exports CSV stats plus ready-to-share plots.
Mechanistic interpretability tool to detect induction heads in GPT-2 using TransformerLens
Add a description, image, and links to the attention-heads topic page so that developers can more easily learn about it.
To associate your repository with the attention-heads topic, visit your repo's landing page and select "manage topics."