Rendered at 19:05:57 GMT+0000 (Coordinated Universal Time) with Cloudflare Workers.
_jayhack_ 45 minutes ago [-]
Shocking to me how little interest the general public has in mechinterp given the alien capabilities demonstrated by LLMs - this and the subsequent transformer-circuits.pub publications will be seen as classic, foundational work in a few years
myself248 4 hours ago [-]
Nothing about B-H curves or magnetization current, must be that other kind of transformer, in that other kind of circuit.
greenbit 1 hours ago [-]
Someone should step up and rectify that, there's a lot of potential for confusion.
ziofill 1 hours ago [-]
This comment is hysteresical
Bluestein 51 minutes ago [-]
I find it polarizing.-
thatspartan 38 minutes ago [-]
It can be quite eddyfying
amelius 4 hours ago [-]
> A previous project, the Distill Circuits thread, has attempted to reverse engineer vision models, but so far there hasn’t been a comparable project for transformers or language models.
How successful was that?
- https://distill.pub/2021/distill-hiatus/
... and, here we are.-
https://youtu.be/KV5gbOmHbjU?is=GSiv0rSHBocNdU8l
Also long, but it's detailed and complicated, so there is no escaping that.