korrents

Our readingWhat is a korrent?

Matching vocabularies across model families for soft distillation is super hacky, so most people just do hard distillation.

Drawn from what Sebastian Raschka said

LLMs

What Sebastian Raschka actually said

  1. Sebastian Raschka AI research engineer working on large language models
    Of course, there are hacks and workaround you can kind of synchronize these vocabularies and kind of make them match up and stuff like that. But it's it's super hacky and tricky. And so most people in practice, they just do hard distillation because it's just easier than doing this.

    Watch from 0:08:47 plays here

    ↗Build A Reasoning Model From Scratch 8: Distillationyoutube.com

    17 Oct 2026 · video · 1h 21m · spoken · machine transcript

    All 23 korrents from this recordingthis one is 3rd

Added to korrents 9 Nov 2026 · How quotes work · Something wrong? Tell us

Do you hold this korrent?Do you also believe this?

I hold this too

Related korrents

Shared subject first, then similar wording. They may agree or disagree.

See this claim on the map