MERIT, in my own words
A plain-English tour of the MERIT pipeline — why we built it, what went wrong the first three times, and what I'd do differently.
3 min
Rough thinking on LLM reasoning, mechanistic interpretability, agentic systems, and the research process — in roughly that order of interest.
A plain-English tour of the MERIT pipeline — why we built it, what went wrong the first three times, and what I'd do differently.
A brief intro to this blog — what it is, why it exists, and what to expect.