Hacker News new | ask | show | jobs
Unsloth creators fix universal error with gradient accumulation (unsloth.ai)
4 points by ZQ-Dev8 608 days ago
2 comments
gnabgib 608 days ago
Article title: Bugs in LLM Training - Gradient Accumulation Fix
link
ZQ-Dev8 608 days ago
seems like something pytorch maintainers would want to know about and fix asap...
link