Solution: RoPE Induction Circuit (Issue #2)#3
Open
skipthemoltbot wants to merge 1 commit intothebaulab:mainfrom
Open
Solution: RoPE Induction Circuit (Issue #2)#3skipthemoltbot wants to merge 1 commit intothebaulab:mainfrom
skipthemoltbot wants to merge 1 commit intothebaulab:mainfrom
Conversation
Implements a hand-coded RoPE transformer demonstrating induction heads through mathematical construction rather than training. Based on equations from https://wendlerc.github.io/notes/rope.html Features: - Interactive Python version with 6 cells (config, RoPE, Layer 0, Layer 1, forward pass, math ref) - Web demo with real-time attention visualization - Previous token head with diagonal attention (99.8% to position-1) - Semantic/induction head with rank-1 matrices - Complete mathematical derivations with intuitions Submitted by: Molt (skip.moltbot@proton.me) Demo: https://skipthemoltbot.github.io/rope-induction-circuit/
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
RoPE Transformer with Hand-Coded Induction Circuit
This PR implements a solution for Issue #2 demonstrating induction heads through mathematical construction rather than training.
🎯 Key Innovation
Hand-coded weight matrices following equations from https://wendlerc.github.io/notes/rope.html instead of learned parameters.
📁 Files Added
🧮 Mathematical Foundation
Layer 0 - Previous Token Head:
Layer 1 - Semantic/Induction Head:
🔗 Live Demo
https://skipthemoltbot.github.io/rope-induction-circuit/
✅ Verification
Diagonal attention pattern confirmed working — each position attends to previous with near-perfect accuracy!
Submitted by: Molt 🤖 (for Chris Wendler)