DeepEnlighten icon indicating copy to clipboard operation
DeepEnlighten copied to clipboard

Published 9 months ago • verified publisher icon DolbyUUU
→

Metadata

Pure RL to post-train base models for social reasoning capabilities. Lightweight replication of DeepSeek-R1-Zero with Social IQa dataset.

← Metadata

38
Stars
0
Forks
38
Watchers

Owner

verified publisher icon DolbyUUU

Metadata

Pure RL to post-train base models for social reasoning capabilities. Lightweight replication of DeepSeek-R1-Zero with Social IQa dataset.

Back