AI ALIGNMENT FORUM
AF

254
Monthly Algorithmic Problems in Mech Interp

Monthly Algorithmic Problems in Mech Interp

Sep 25, 2023 by CallumMcDougall

This sequence collects all the monthly algorithmic problems in mechanistic interpretability, which have been created to go alongside the first chapter of ARENA material.

An invite to the Slack group can be found here: https://join.slack.com/t/arena-la82367/shared_invite/zt-1uvoagohe-JUv9xB7Vr143pdx1UBPrzQ.

14Mech Interp Challenge: August - Deciphering the First Unique Character Model
CallumMcDougall
2y
1
14Mech Interp Challenge: September - Deciphering the Addition Model
CallumMcDougall
2y
0
7Mech Interp Challenge: October - Deciphering the Sorted List Model
CallumMcDougall
2y
0
6Mech Interp Challenge: November - Deciphering the Cumulative Sum Model
CallumMcDougall
2y
0
10Mech Interp Challenge: January - Deciphering the Caesar Cipher Model
CallumMcDougall
2y
0