PaTaRM PaTaRM is a Generative Reward Model (GRM) for RLHF alignment. AIJian/PaTaRM-8B Text Generation • 0.5B • Updated Apr 2 • 106 AIJian/PaTaRM-data Preview • Updated Apr 1 • 30 AIJian/PaTaRM-14B Text Generation • 0.5B • Updated Apr 1 • 88 AIJian/PaTaRM Updated Apr 1
PaTaRM PaTaRM is a Generative Reward Model (GRM) for RLHF alignment. AIJian/PaTaRM-8B Text Generation • 0.5B • Updated Apr 2 • 106 AIJian/PaTaRM-data Preview • Updated Apr 1 • 30 AIJian/PaTaRM-14B Text Generation • 0.5B • Updated Apr 1 • 88 AIJian/PaTaRM Updated Apr 1