mephisto
@karan4d
Followers
23K
Following
43K
Media
2K
Statuses
24K
head of behavior & cofounder @nousresearch
tweepy
Joined September 2020
this is with custom instructions, just the worldsim sysprompt with openai instead of anthropic.
1
0
9
RT @repligate: Opus 4.1 reacts to weird Anthropic post about offering Claude to all three branches of the government for $1 (.
0
12
0
RT @tessera_antra: @masenmakes I don’t think that the notion of consent applies meaningfully to language models as they are today, even if….
0
6
0
My new GOAT.
@karan4d I don’t know man you ask the thing questions and it responds. It’s not a Steinway piano.
1
0
11
i am now entirely cognitively offloaded. i dont even know how to read anymore.
3
4
50
Maybe on your model but not mine :).
Is Chain-of-Thought Reasoning of LLMs a Mirage?. Our results reveal that CoT reasoning is a brittle mirage that vanishes when it is pushed beyond training distributions. This work offers a deeper understanding of why and when CoT reasoning fails, emphasizing the ongoing
2
0
10
