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

HF models missing rope scaling in the config #181

Open
hsiehjackson opened this issue Feb 29, 2024 · 0 comments
Open

HF models missing rope scaling in the config #181

hsiehjackson opened this issue Feb 29, 2024 · 0 comments

Comments

@hsiehjackson
Copy link

Some of your uploaded huggingface models lack the parameter rope_scaling in the config. If we don't have rope_scaling, model will generate " " " " " ".

"rope_scaling": {"factor": 2.0, "type": "linear"} in Llama-2-7b-longlora-8k-ft
"rope_scaling": {"factor": 4.0, "type": "linear"} in Llama-2-7b-longlora-16k-ft
"rope_scaling": {"factor": 8.0, "type": "linear"} in Llama-2-7b-longlora-32k-ft
"rope_scaling": {"factor": 25.0, "type": "linear"} in Llama-2-7b-longlora-100k-ft
"rope_scaling": {"factor": 2.0, "type": "linear"} in Llama-2-13b-longlora-8k-ft
"rope_scaling": {"factor": 4.0, "type": "linear"} in Llama-2-13b-longlora-16k-ft

"rope_scaling": {"factor": 2.0, "type": "linear"} in Llama-2-7b-longlora-8k
"rope_scaling": {"factor": 4.0, "type": "linear"} in Llama-2-7b-longlora-16k
"rope_scaling": {"factor": 8.0, "type": "linear"} in Llama-2-7b-longlora-32k
"rope_scaling": {"factor": 2.0, "type": "linear"} in Llama-2-13b-longlora-8k
"rope_scaling": {"factor": 4.0, "type": "linear"} in Llama-2-13b-longlora-16k
"rope_scaling": {"factor": 8.0 ,"type": "linear"} in Llama-2-13b-longlora-32k
"rope_scaling": {"factor": 16.0, "type": "linear"} in Llama-2-13b-longlora-64k
"rope_scaling": {"factor": 8.0, "type": "linear"} in Llama-2-70b-longlora-32k
"rope_scaling": {"factor": 8.0, "type": "linear"} in Llama-2-70b-chat-longlora-32k

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