# @voooooogel — 2025-02-01

♥8 ↻0 · https://x.com/voooooogel/status/1885519980034482481

@max_paperclips i think the ideal would be to seed a few structures and then hope R1-Zero style that the model can generalize beyond that, into some kind of "meta-reasoning". the humanizing SFT they did to R1 seems to have actually clamped down on that creativity, sadly.

tags: author:voooooogel, kind:tweet, model:deepseek-r1, model:deepseek-r1-zero, on:deepseek-r1-zero, year:2025
cited on: _dossiers/deepseek-r1-zero.md, deepseek-r1-zero