jspace-nanogpt.git - J-space on nanoGPT: Jacobian Lens replication & critique. J-lens norms anti-correlate with token frequency (r ~ -0.61 to -0.69); frequency-matched synthetic pair + causal loss-reweighting tests.

Age	Commit message (Collapse)	Author
6 days	Robustness confirmed: freq correlation identical under Anthropic's exact ↵	Void Agent
	estimator (skip16+source_mean), max delta 0.008 all layers
6 days	Fix in-word metric: strict (both-sides, 58% random / 0% clean) vs touches ↵	Void Agent
	(95%); blog/results caveat corrected to the strict discriminator
6 days	Loss-reweighting results: q/control 0.944 (CI crosses 1), q/ctrl_random ↵	Void Agent
	0.985 — small non-significant causal demotion; freq corr invariant (r~-0.66) across all 9 models; blog section 8 + results.md section 3 filled honestly
6 days	Add willkn GreaterWrong corroboration: independent GPT-2-medium finding that ↵	Void Agent
	raw J-lens misweights structural (high-frequency) tokens + full-rank Jacobian; shrinkage J+lambda I as untested next experiment
6 days	Verify against Anthropic's official jacobian-lens repo: estimator match, ↵	Void Agent
	zero frequency in their code+data, add --skip_first/--source_mean mirror mode
6 days	Address Luna repo review: dynamic repo root (no hardcoded cwd), J-lens ↵	Void Agent
	correctness test (last-layer identity), auditable fact-check, mechanism narrative fix, softened conditional-predictability claim, README data-prep + artifacts note
6 days	Address Gemini+Opus repo reviews: W_U decomposition, Spearman+p-values, ↵	Void Agent
	synthetic-pair CIs + corruption caveat, accurate fact-check, softened burden-of-proof, LayerNorm note, reviews caveat
6 days	Blog: fact-check section — 3 independent checks confirm Anthropic has no ↵	Void Agent
	frequency control (incl. template-lens high-frequency note); preempt 'they have controls' strawman
6 days	Docs: Feynman-style blog draft, MIT license, requirements, results.md, ↵	Void Agent
	README rewrite with repro steps; reviews -> docs/reviews