Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

add force_greedy_sample #704

Open
wants to merge 9 commits into
base: mlperf_features
Choose a base branch
from

Conversation

jikunshang
Copy link

@jikunshang jikunshang commented Jan 20, 2025

No description provided.

xuechendi and others added 7 commits December 16, 2024 19:30
Signed-off-by: Chendi Xue <chendi.xue@intel.com>
Signed-off-by: Kunshang Ji <kunshang.ji@intel.com>
Signed-off-by: Chendi Xue <chendi.xue@intel.com>
…roject#630)

```
VLLM_TP_SPLIT_SIZE_BY_SEQ=2
VLLM_TP_SPLIT_SIZE_BY_BATCH=2
```

Signed-off-by: Kunshang Ji <kunshang.ji@intel.com>
Signed-off-by: Chendi Xue <chendi.xue@intel.com>
Co-authored-by: Kunshang Ji <kunshang.ji@intel.com>
Signed-off-by: Chendi Xue <chendi.xue@intel.com>
https://jira.habana-labs.com/browse/SW-212933
```
VLLM_PIPELINED_PA=true \
VLLM_SOFTMAX_CONST_NORM=true
```

Signed-off-by: Chendi Xue <chendi.xue@intel.com>
Enable splitting qkv and gate_up into separate columnparallellinear
layers, instead of overwriting qkvparallellinear and
mergedcolumnparallellinear layers.

---------

Signed-off-by: Chendi Xue <chendi.xue@intel.com>
Co-authored-by: Kunshang Ji <kunshang.ji@intel.com>
Co-authored-by: Chendi Xue <chendi.xue@intel.com>
Co-authored-by: Barak Goldberg <bgoldberg@habana.ai>
Co-authored-by: Nir David <ndavid@habana.ai>
Signed-off-by: Kunshang <kunshangli@habana.ai>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants