Research / Interpretability
Anthropic research article on Claude interpretability, presenting evidence for a J-space and a global-workspace-like mechanism in language models.
The page is a substantial, clearly described AI research post from Anthropic with enough visible detail for a public listing. It explains the core idea, names the technique used, summarizes multiple findings about Claude’s internal representations, and links the post to a paper, code repository, and demo.
The crawl does not show the linked paper, code repository, or demo details, so a visitor cannot tell the exact methods, reproducibility setup, or how to use the released materials from this page alone.
Last evaluated 10 days ago. Current rank #271. Up 2 spots in the rankings.
An interactive dashboard that analyzes New York Times coverage since 2000 using the NYT Archive API, with views for reporters, beats, sections, subjects, geography, obituaries, and corrections.
An open benchmark and leaderboard for AI CAD agents, with 308 prompts across 20 categories and layered scoring for geometry, engineering, manufacturability, and cognition.
A research article from Applied Compute on how agentic, tool-using workloads differ from traditional LLM benchmarks, with production observations, workload profiles, and an open-source harness for replaying traces.
UnderstandDocs is a document analysis tool that summarizes pasted text or uploaded files, flags risks, extracts important dates, and simplifies dense language. The page shows a working analysis form, supported file types, a privacy/no-storage claim, and an example output for a tenancy agreement.
