Article
Native FP8 Mixed Precision Training for Ling 2.0, Open Sourced！
By
•
•
4Qian Zhao
im0qianqian
AI & ML interests
multi-hop reading comprehension, commonsense reasoning
Recent Activity
authored
a paper
10 days ago
Every Attention Matters: An Efficient Hybrid Architecture for
Long-Context Reasoning
upvoted
a
paper
10 days ago
Every Step Evolves: Scaling Reinforcement Learning for Trillion-Scale
Thinking Model
Organizations
None yet
published
an
article
about 2 months ago