Hi, I am trying to use BioGPT large to finetune my dataset and I have replaced the --arch transformer_lm_prompt_biogpt_large in the train.sh file. But I am still getting the mismatch error. I believe this is due to the wring size of vocab but I don't know where to change it as I was not able to find the right file for it.
Hi, I am trying to use BioGPT large to finetune my dataset and I have replaced the --arch transformer_lm_prompt_biogpt_large in the train.sh file. But I am still getting the mismatch error. I believe this is due to the wring size of vocab but I don't know where to change it as I was not able to find the right file for it.