rename BIGDL_OPTIMIZE_LM_HEAD to IPEX_LLM_LAST_LM_HEAD and add qwen2 (#11418)
This commit is contained in:
parent
75f836f288
commit
9e4ee61737
1 changed files with 2 additions and 2 deletions
|
|
@ -328,8 +328,8 @@ def _replace_with_low_bit_linear(model, qtype, modules_to_not_convert=None,
|
|||
optimize_lm_head = False
|
||||
if is_lm_head(name, model_config, out_features):
|
||||
model_type = getattr(model_config, "model_type", None)
|
||||
if model_type in ["gptj", "llama"] and os.environ.get("BIGDL_OPTIMIZE_LM_HEAD",
|
||||
None) == "1":
|
||||
if model_type in ["gptj", "llama", "qwen2"] and \
|
||||
os.environ.get("IPEX_LLM_LAST_LM_HEAD", None) == "1":
|
||||
optimize_lm_head = True
|
||||
with init_empty_weights():
|
||||
new_linear = None
|
||||
|
|
|
|||
Loading…
Reference in a new issue