Been making #NoriHoh# practice her fighting stance lately. I think she's doing well so far... I guess you can all her Ip-Man now.. sorry I mean #bibman#
Also take that weekdays! Because it's #TGIF#! Wattaa!
#Hohpa# #babyboxer# #boxinglife# #mmatraining# #babyfighter#
New Anthropic Fellows research: Model Spec Midtraining (MSM).
Standard alignment methods train AIs on examples of desired behavior. But this can fail to generalize to new situations.
MSM addresses this by first teaching AIs how we would like them to generalize and why.