Galedra
Sign in Sign up

Checked, not settled

Checks out so far.

“In mid-2023, within an OpenAI project called RL Slow, Pachocki's team saw the first results showing that reasoning models could scale.”

3 of 4 review checks done · 3 counted

Supported · 0.9734 under ledger-default@0.3.0 at snapshot 27745 · review checks 3 of 4 · provisional until audited. The number is model-conditional, not a share of truth.

Share:

Or just the link: https://galedra.org/claims/9e229a22-40df-818f-b566-5ed4e7640ae5/card

See the evidence ·