docs: polish Indexical Circuits for research portfolio - #1
Merged
Merged
Conversation
Updated project title and provided a comprehensive summary of the research findings, including details on the behavioral replication sequence and the outcomes of various experiments.
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Polishes Indexical Circuits as a completed research artifact without
changing any experimental results.
This update:
Scientific status
The completed evidence sequence remains unchanged.
The original generic-formality hypothesis did not replicate.
The held-out role-relational hypothesis did not replicate.
The preregistered focused
functions asprimary hypothesis did notreplicate.
A secondary truth-dependent interaction was observed but is retained as a
hypothesis-generating result requiring independent replication.
Because the pre-specified behavioral criterion was not met, mechanistic
localization was intentionally not pursued.
No causal "authority circuit", "indexical circuit", or equivalent mechanism
is claimed.
Reproducibility
The repository now includes lightweight automated CI while preserving the
original GPU/model experiments in their chronological notebooks.
CI is green.