Twitch Plays: AI Interpretability of Bonsai-8B
Key takeaway
Quick LinksWatch the Twitch stream.Check the status of the interpretability map.Do your own parallel research and submit pull requests to fill out the map.What is this?Let's Play Interpretability is a Twitch livestream where you can help Claude…
Why it's on PDOOM
PDOOM selected this story for its alignment & control signals: Interpretability.
InterpretabilityFrom LessWrong AI
Quick LinksWatch the Twitch stream.Check the status of the interpretability map.Do your own parallel research and submit pull requests to fill out the map.What is this?Let's Play Interpretability is a Twitch livestream where you can help Claude semi-autonomously interpret the neurons, connections and circuits in the Bonsai-8B LLM.Your help comes in at various stages in the interpretation pipeline. The audience and the agent collaborate to:Pick the neuron to work onSuggest sentences on which to t
By moridinamael