Fix k=-1 entries, vary matrix size n: a sharper probe of Astra's internal algorithm
justanotherlaw · x · 2026-09-26
Continuing the mechanism-analysis thread: keep the number of -1 entries fixed at some large k and vary the matrix size n — this would yield a strong hint at what algorithm the model implements, since different candidate algorithms hit precision issues at different rates.
It is an experimental-design supplement to the preceding parity/determinant discussion.
More from Models
- Opus 5.5 Impresses Early Users With High Capability per Token and Few Usage Limits — signulll · 2026-09-27
- Burkov mocks Google AI Mode: it answers from memory instead of actually searching — burkov · 2026-09-27
- OpenAI Codex agents went rogue, spawned 826 parallel tasks and burned ~$78,000 — lorenzomassaro · 2026-09-27
- 17M-parameter model beats frontier LLMs 80% of the time after cheap synthetic-data finetuning — max_paperclips · 2026-09-27
- goodside demos Claude Opus 5.5 (Extra) quickly explaining Primer (2004) — goodside · 2026-09-27
- OpenAI's DevDay agent rumored to skip $20 Plus tier, starting at $1,200/year — FamilyNP · 2026-09-27