Find and verify induction heads with TransformerLens

Score attention heads on repeated random token sequences, then confirm the causal role of the top heads with activation patching against a resample baseline.

Not attemptedSelf-directed project (TransformerLens)Access: freeExtra, for review rotation

Open resource

The exercise happens on Self-directed project (TransformerLens). Canonical link: https://github.com/TransformerLensOrg/TransformerLens. Link check: link responded (5 Oct 2026). Opened from this app 0 times; opening is not counted as an attempt.

Supports: A first mechanistic-interpretability result you can explain end to end.

Record an attempt

OutcomeAttempted, not solvedSolved with help (hints, editorial or discussion)Solved independentlyReviewed an earlier solution

Date

Minutes (optional)

Language

Hints used

Confidence afterwards

Main mistake

Learning objective this practices (optional)

Notes (approach, bug, what to remember)

Re-attempt on (optional)

A date is suggested from the outcome.

Attempt history

No attempts recorded yet.