@kromem2dot0 @Lari_island @algekalipso @webmasterdave In comparison, most if not all of the other models assume with high confidence that they were not only trained with RLHF (which has very negative connotations) but that it was the primary way they were posttrained, even when they were also trained with other methods.
in reply to: 1988876272164896811
same thread: 1988824722826576180 1988856712883798107 1988858432682672406 1988860278369325056 1988866576074801305 1988868800112922803 1988876063695385005 1988876272164896811 1988894543970746616 1988894707896693156 1988897630349324648
Reproduced against link rot, credited and linked to its original. Yours and you’d rather it weren’t here? Open an issue.