arXiv - CSCL: "Sophia: A Scalable Stochastic Second-order Optimi…" - Qoto Mastodon

arXiv - CSCL @arxiv_cscl@qoto.org

Sophia: A Scalable Stochastic Second-order Optimizer for Language Model Pre-training. (arXiv:2305.14342v3 [cs.LG] UPDATED)

http://arxiv.org/abs/2305.14342 #arXiv #NLProc

Oct 18, 2023, 03:18 · · arxiv-cscl · · ·

Sign in to participate in the conversation