News for hackers. Stuff that matters.

A comment on Artificial Insanity. Parent comment

Re:Artificial Insanity by verdverm on Sunday October 11, 2026 @06:37PM (#50048676)
the most widely used API surfaces are already shaped around chat, it's different and most analogies will fall short in some way
[ Parent | on HN ]
Re:Artificial Insanity by cluckindan on Sunday October 11, 2026 @07:06PM (#50048988)
Neither a chat or an API are an LLM.
[ Parent | on HN ]
Re:Artificial Insanity by verdverm on Sunday October 11, 2026 @07:09PM (#50049010)
LLM's have chat templates that sit at the lowest levels of how we feed tokens to them. It's quite integral to how they are trained to process tokens and rises up through the tech stack.

Go to a huggingface repo, look at the files, they have a literal chat_template.jinja

[ Parent | on HN ]
Re:Artificial Insanity by cluckindan on Tuesday October 13, 2026 @03:49PM (#50077610)
Are you saying an engine with a steering rack is a car?
[ Parent | on HN ]
Re:Artificial Insanity by tptacek on Sunday October 11, 2026 @07:29PM (#50049198)
By that standard Gebru was also not writing about LLMs in the Stochastic Parrot paper.
[ Parent | on HN ]