good first issue
仓库指标
- Star
- (6,010 star)
- PR 合并指标
- (平均合并 1天 11小时) (30 天内合并 31 个 PR)
描述
Currently the context parallelism's training time is like 2 times when we use cp=2 than cp=1, we should try to optimize this for long sequence, maybe reduce the all gather entropy and logp and calculate the local loss to reduce the numbers of communication