Content
@
0 reply
0 recast
0 reaction
@
0 reply
0 recast
0 reaction
eric siu π
@randomishwalk
ngl code-llama has been better even claude-2 sometimes gpt-4 is not as good as it used to be
4 replies
0 recast
2 reactions
Keith Axline
@kaxline
Just noticed this too. Weird that itβs getting worse. What would make that happen? I thought it was ChatGPT specifically and they were making adjustments behind the scenes that were degrading it
1 reply
0 recast
0 reaction
eric siu π
@randomishwalk
RLHF (leading hypothesis i think) and other stuff
1 reply
0 recast
0 reaction