You can't accidentally retrain a model to use a different tokenizer. It changes the input vectors to the model.
replyI appreciate you saying that, i think sometimes with ai conversations i wade into them without knowing the precise definitions of the terms, I'll try to be more careful next time. Thank you.
reply