Skip to content

Commit

Permalink
Update README.md
Browse files Browse the repository at this point in the history
  • Loading branch information
Tlntin committed Sep 16, 2023
1 parent 14f0746 commit 242703d
Showing 1 changed file with 2 additions and 2 deletions.
4 changes: 2 additions & 2 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -182,12 +182,12 @@
4. 开始编译trt_engine(未完待续)
- 普通版 (调用第三步编译的rmsnorm)
```bash
python3 build.py --use_smooth_quan --use_rmsnorm_quantization_plugin=fp16
python3 build.py --use_smooth_quan --use_rmsnorm_quantization_plugin=float16
```

- 升级版(理论上速度更快一些)
```bash
python3 build.py --use_smooth_quan --use_rmsnorm_quantization_plugin=fp16 \
python3 build.py --use_smooth_quan --use_rmsnorm_quantization_plugin=float16 \
--per_token --per_channel
```

Expand Down

0 comments on commit 242703d

Please sign in to comment.