423A35C7/LLaMA-Factory
1
0
Fork 0
mirror of https://github.com/hiyouga/LLaMA-Factory.git synced 2025-11-06 02:42:15 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix valuehead model

Browse Source 
Former-commit-id: 9f628debb6510f2d1c91b00f121a721ab5d648e9
This commit is contained in:
hiyouga 2023-12-14 20:15:20 +08:00
parent c9ff152cef
commit 511f3f68e2
1 changed files with 3 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
src/llmtuner/model/loader.py
View File

@ -203,6 +203,9 @@ def load_model_and_tokenizer(
# Prepare model with valuehead for RLHF # Prepare model with valuehead for RLHF
if add_valuehead: if add_valuehead:
model: "AutoModelForCausalLMWithValueHead" = AutoModelForCausalLMWithValueHead.from_pretrained(model) model: "AutoModelForCausalLMWithValueHead" = AutoModelForCausalLMWithValueHead.from_pretrained(model)
def get_input_embeddings(self: "AutoModelForCausalLMWithValueHead") -> torch.nn.Module:
return self.pretrained_model.get_input_embeddings()
setattr(model, "get_input_embeddings", MethodType(get_input_embeddings, model))
ignore_modules = [name for name, _ in model.named_parameters() if "pretrained_model" in name] ignore_modules = [name for name, _ in model.named_parameters() if "pretrained_model" in name]
setattr(model, "_keys_to_ignore_on_save", ignore_modules) setattr(model, "_keys_to_ignore_on_save", ignore_modules)
setattr(model, "tie_weights", MethodType(lambda _: None, model)) # use empty method setattr(model, "tie_weights", MethodType(lambda _: None, model)) # use empty method