in retrospect, we should have done more to temper expectations and highlight the difference between the method/architecture/open-source code and the actual pretrained weights, which we def want ppl to play with, but come with absolutely no guarantees.

(disclaimer: opinions are my own)

we used language that implied that the pretrained Octo weights would definitely actually be useful for YOU in YOUR lab. in reality, the real world is much bigger than OXE and some carefully-tuned evals. end-to-end robot learning is not even at the GPT-2 level

super happy to hear the first one, since we worked so hard on open-sourcing! the last one is the one that I'm less confident about -- it's very hard to beat a well-tuned from-scratch baseline. but if it works for you, then perhaps I've been too pessimistic!

don't worry I wasn't insulted :)

interestingly, this is the second time that public dunking on Octo has informed me of someone who actually does use it (like @tomdupuis.bsky.social), revising my opinion of its utility upwards rather than downwards.

(the first time I was the one doing the dunking)

yeah we 100% oversold Octo

tokenization is inductive bias

same, I've been trying to force myself to use btop but something about htop just feels... comfy... maybe it's just familiarity bias