LLaVA
LLaVA copied to clipboard
[Question] How to fix the maximum sequence length of the model
Question
How to fix the maximum sequence length of the model? When I using finetuned model to test my own data, I found error: Token indices sequence length is longer than the specified maximum sequence length for this model(2199>2048).
Do I need to fix weights in this script?
Thanks!!!
same doubt
@crazysal Do you solve it?