Skip to content

GitLab

Unverified Commit 3102a8e4 authored by MaYongQing's avatar MaYongQing Committed by GitHub
Browse files

fix: vllm lora (#3132)

parent cf631de0
Show whitespace changes
Inline Side-by-side
Showing with 3 additions and 0 deletions
lm_eval/models/vllm_causallms.py
View file @ 3102a8e4
...@@ -134,6 +134,7 @@ class VLLM(TemplateLM): ...@@ -134,6 +134,7 @@ class VLLM(TemplateLM):
data_parallel_size: int = 1, data_parallel_size: int = 1,
lora_local_path: str = None, lora_local_path: str = None,
enable_thinking: bool = False, enable_thinking: bool = False,
max_lora_rank: int = 16,
**kwargs, **kwargs,
): ):
super().__init__() super().__init__()
...@@ -167,6 +168,8 @@ class VLLM(TemplateLM): ...@@ -167,6 +168,8 @@ class VLLM(TemplateLM):
"quantization": quantization, "quantization": quantization,
"seed": int(seed), "seed": int(seed),
"device": str(device), "device": str(device),
"enable_lora": True if lora_local_path else False,
"max_lora_rank": int(max_lora_rank),
} }
self.model_args.update(kwargs) self.model_args.update(kwargs)
self.batch_size = ( self.batch_size = (
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment