Jacobian-Lens interpretability primitive for open-weight LLMs (Anthropic 2026-07-06, 'Verbalizable Representations Form a Global Workspace'). Reads the model's functional workspace at inference time — lens_l(h)=unembed(J_l·h) — into workspace tokens, laye
@metaharness/[email protected] low health (65/100) — consider alternatives
Get this data programmatically — free, no authentication.
curl https://depscope.dev/api/check/npm/@metaharness/workspace-lensFirst published · 2026-07-07T02:49:11.293Z
Last updated · 2026-08-02T16:41:33.944Z