Hacker News
new
past
comments
ask
show
jobs
points
by
tjungblut
5 hours ago
|
comments
by
7moritz7
3 hours ago
|
[-]
These are sota models, not open source 7b parameter ones. They've put lots of effort into preventing prompt injections during the agentic reinforcement learning
reply