23 mar. 2023

Playing with the supposedly GPT3 quality LLaMA and Alpaca models really highlights how there is a quality level beyond which these things become suddenly useful.

The responses from Alpaca are "mostly" sort of there, but unreliable in my experience.

ChatGPT is just so much more confidence inspiring to use, because it is just the other side of the "mostly coherent sentences, but not quite what I wanted" to "more or less what I wanted, most of the time".

Vous voulez savoir quand je poste du contenu sur mon blog ? Il suffit de vous inscrire gratuitement à un agrégateur RSS (Feedly, NewsBlur, Inoreader, …) et d'ajouter à vos flux (ou pour vous abonner à tous les sujets). On n'a pas besoin de newsletters, pas besoin de Twitter, le RSS existe toujours.