Fine-tune ChatGPT w/ in-context learning ICL - Chain of Thought, AMA, reasoning & acting: ReAct

[@nextlesson]

This was great, thank you for putting this together. Could you make a future video where you go through more practical examples on how to apply ReACT and ICL on the prompt?

[@code4AI]

Sure thing! Vids already online: kzbin.info/www/bejne/kJDCnJxvqrhpmc0 kzbin.info/www/bejne/sJbMc3tqbZl7aMk kzbin.info/www/bejne/sHjZaKaKhN2Lras

[@theycallmeken]

Would you please consider releasing your slides under Creative Commons 0? Thank you !

[@theshrubberer]

thanks. Question ...does the ICL achieved by the prompt or prefix tuning persist beyond the current context and become part of the model going forward? If so the models are infinitely extensible in this way? The model grows by the cumulative ICL input of all users?

[@vmfox2152]

"...patched by Microsoft..." 🤣😂🤣 Made my day!

[@riser9644]

Prefix vs Lora vs adapter, aren't they all doing the same thing

[@progamer1196]

I always hear "Hello come unity"

[@code4AI]

Sometime someone hears what he/she/they want to hear ...

[@WillMcCartneyAI]

is there a discord or something for the community?

[@code4AI]

Reddit ...

[@matmathugamage2463]

@@code4AI what is the subreddit called? thank you!

[@fraternitas5117]

@@code4AI yuck