You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Downloading llama models from hugging face requires llama to previously authorise your account. it's two weeks since I submitted the request, but it is still pending. So, I downloaded the models from the Lllama website, but they come in a different format. I wonder if there is a way to convert them to hugging face format in order to use them with litgpt.
michele
The text was updated successfully, but these errors were encountered:
Hi there. Sorry to hear that the approval is such a hassle. It seems there must be an issue on their website as it should usually be close to immediate.
I wish I had some code to share on the conversion of the weights from the Llama website, but unfortunately we don't support that at the moment. I think our old project Lit-Llama had such a conversion script, but that was only for the original Llama model: https://github.com/Lightning-AI/lit-llama/blob/main/scripts/convert_checkpoint.py
Hi,
Downloading llama models from hugging face requires llama to previously authorise your account. it's two weeks since I submitted the request, but it is still pending. So, I downloaded the models from the Lllama website, but they come in a different format. I wonder if there is a way to convert them to hugging face format in order to use them with litgpt.
michele
The text was updated successfully, but these errors were encountered: