datasets Collection Reasoning Core ◉ Pre-generated symbolic reasoning data, from pre-training pile to post-training environments • 5 items • Updated Mar 3 • 4
view article Article Shipping a Trillion Parameters With a Hub Bucket: Delta Weight Sync in TRL +6 aminediroHF, qgallouedec, kashif, lewtun, edbeeching, albertvillanova, lvwerra, sergiopaniego • 15 days ago • 41