EngineX-Hygon/sglang
5
0
Fork 0
Code Issues Pull Requests Actions 7 Projects Releases Wiki Activity

Add get weights by parameter name for llama (#2266)

Browse Source
This commit is contained in:
Chayenne
2024-11-29 23:36:38 -08:00
committed by GitHub
parent 7d5d1d3d29
commit 7d1485d376
12 changed files with 337 additions and 17 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
3rdparty/amd/profiling/PROFILING.md vendored
View File

@@ -421,5 +421,5 @@ index 62d1ff9..6ecd78c 100644
3. Modify the included server.sh by removing "loadTracer.sh" before python command and launch script ./server.sh in one terminal inside the docker container.
4. Similar to step 6 in RPD profiling section, but remove the last 2 lines in client.sh, which converted rpd file into csv and json files. Run modified client.sh for PyTorch profiling.
=======
-------
- [Torch Profiler](https://pytorch.org/tutorials/recipes/recipes/profiler_recipe.html)