From 781a07b0c52c62657a8030eec8bbab76d26fa90f Mon Sep 17 00:00:00 2001 From: Void Agent Date: Sun, 2 Aug 2026 14:54:17 +0100 Subject: Add willkn GreaterWrong corroboration: independent GPT-2-medium finding that raw J-lens misweights structural (high-frequency) tokens + full-rank Jacobian; shrinkage J+lambda I as untested next experiment --- docs/blog-jlens-frequency.md | 14 ++++++++++++++ 1 file changed, 14 insertions(+) (limited to 'docs/blog-jlens-frequency.md') diff --git a/docs/blog-jlens-frequency.md b/docs/blog-jlens-frequency.md index e73a2fe..7e01de0 100644 --- a/docs/blog-jlens-frequency.md +++ b/docs/blog-jlens-frequency.md @@ -198,6 +198,20 @@ either, and its one acknowledgment of high-frequency-token trouble was in a separate method they chose not to use. Any "privileged subspace" interpretation needs a frequency control first. +We are also not alone in noticing the raw lens is distorted by token +statistics. An independent research-engineer analysis of the same paper +(willkn, "Anthropic's J-Lens: A Research Engineer's Analysis", GreaterWrong, +24 Jul 2026), working on GPT-2-medium (355M — thirty times our model), found +that the raw fitted Jacobian "misweights structural tokens (grammar, +punctuation) over semantic content" — its dominant spectral channels carry +~10x the gain of the residual pathway. Structural tokens are the high-frequency +tokens. They also found the Jacobian essentially full-rank (562-858 dimensions +for 90% of spectral variance), matching our toy-scale rank result. Their fix is +a single-parameter shrinkage regularizer (J + λI) that restores next-token +faithfulness — and Anthropic's released fitting code applies no such +regularization. Whether shrinkage also removes the frequency correlation is an +experiment we have not run yet; it is a natural next step. + ## 7. But not *only* frequency Now the twist. Correlation is not causation, so we ran a cleaner test. We made -- cgit v1.2.3