frmsaul

• frmsaul 14 Apr 2026 22:52 UTC

  1 point

  0

  in reply to: StanislavKrym’s comment on: You’re gonna need a big­ger boat (bench­mark), METR

  The scaffold result is very cool and surprising. It seems like terminal-bench find pretty big variance across scaffolds. It also seems that cognition sees a lot of variance between different scaffolds.
  
  It’s cool that TH1.1 is so robust to this. Why do you think that is?

• frmsaul 12 Apr 2026 19:04 UTC

  3 points

  0

  on: Why Con­trol Creates Con­flict, and When to Open Instead

  I loved the cartoon.