DeepSeek/profile-data
1
0
Fork 0
mirror of https://github.com/deepseek-ai/profile-data synced 2025-04-03 12:21:12 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Merge pull request #4 from vatlor/main

Browse Source 
fix typo
This commit is contained in:
Chengqi Deng 2025-03-03 18:09:56 +08:00 committed by GitHub
commit e66dec569b
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
README.md
View File

@ -9,7 +9,7 @@ Here, we publicly share profiling data from our training and inference framework
![train](assets/train.jpg)
The training profile data demonstrates our overlapping strategy for a pair of individual forward and backward chunks in [DualPipe](https://github.com/deepseek-ai/dualpipe). Each chunk contains 4 MoE (Mixture of Experts) layers.
The parallel configuration aligns with DeepSeek-V3 pretraining settings: EP64, TP1 with 4K sequence length. And the PP communication is not included during profilng for simplicity.
The parallel configuration aligns with DeepSeek-V3 pretraining settings: EP64, TP1 with 4K sequence length. And the PP communication is not included during profiling for simplicity.
## Inference