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

Cannot load bias from the intervened model #49

Open
ZhongqiangWu opened this issue Feb 12, 2025 · 0 comments
Open

Cannot load bias from the intervened model #49

ZhongqiangWu opened this issue Feb 12, 2025 · 0 comments

Comments

@ZhongqiangWu
Copy link

ZhongqiangWu commented Feb 12, 2025

I am using the ITI method in Qwen model, and it worked when in edit_weight step.
However, when I load the previously saved model , the error occurred as following:

{path_prefix}/validation/results_dump/edited_models_dump//Qwen2.5-14B-Instruct-GPTQ-Int4_seed_42_top_24_heads_alpha_15 were not used when initializing Qwen2ForCausalLM: ['model.layers.17.self_attn.o_proj.bias', 'model.layers.21.self_attn.o_proj.bias', 'model.layers.23.self_attn.o_proj.bias'

The outputs of the original model and the intervened model are identical.

I am using AutoModelForCauselLM to load model. I have been troubled by this error for several days, thank you.

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

No branches or pull requests

1 participant