ayo/ipex-llm
1
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 1

add fused rms optimization for chatglm model (#9256)

Browse source
This commit is contained in:
SONG Ge 2023-10-24 14:40:58 +08:00 committed by GitHub
parent b15656229e
commit bfc1e2d733
1 changed files with 3 additions and 0 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

3
python/llm/src/bigdl/llm/transformers/convert.py
View file

@ -233,6 +233,9 @@ def _optimize_post(model):
convert_forward(model, convert_forward(model,
module.CoreAttention, module.CoreAttention,
core_attn_forward_8eb45c) core_attn_forward_8eb45c)
convert_forward(model,
module.RMSNorm,
llama_rms_norm_forward)
elif hasattr(model.config, 'vocab_size') and model.config.vocab_size == 130528: elif hasattr(model.config, 'vocab_size') and model.config.vocab_size == 130528:
# chatglm-6b # chatglm-6b
modeling_module_name = model.__class__.__module__ modeling_module_name = model.__class__.__module__