Urgent help needed on Llama2 Authentication on Colab

Beginners
1

I’m using AutoTrain-Advanced from Colab. The model I’m using is Llama2 which I have already requested access from Meta and was granted access and HF model indicate the same. When I run the training in colab with Token, it will run for about 30% and finally give and error. The following is a sample of the trace.

Cannot access gated repo for url https://huggingface.co/meta-llama/Llama-2-7b-chat-hf/resolve/main/config.json .
Repo model meta-llama/Llama-2-7b-chat-hf is gated. You must be authenticated to access it.

2

have you logged in using notebook login method

3

thanks for the response, Yes I’m using the following command which includes the authentication username and token.

!autotrain llm --train --project-name test1 --model abhishek/llama-2-7b-hf-small-shards --data-path . --use-peft --lr 2e-4 --auto_find_batch_size --epochs 3 --trainer sft --add_eos_token --text_column text --quantization int8 --model_max_length 2048 --block_size 2048 --merge_adapter –username userName --token xxxxxx

4

i think you must have to login first using

from huggingface_hub import notebook_login

notebook_login() and paste your token and also you must get acess to the model first on llama repo

5

This topic was automatically closed 12 hours after the last reply. New replies are no longer allowed.