v0.4.6
[Feature]: support FP8 communication in DDP, FSDP, Gemini (#5928)
[example] make gpt example directory more clear (#2353)
[hotfix] moe hybrid parallelism benchmark & follow-up fix (#6048)
[Feature] Split cross-entropy computation in SP (#5959)
[misc] refactor launch API and tensor constructor (#5666)
[fp8] Merge feature/fp8_comm to main branch of Colossalai (#6016)
[example]add gpt2 benchmark example script. (#5295)
[devops] remove post commit ci (#5566)