3 ms·

I wanted to better understand the potential impact for the tokenizer change from 4.6 and 4.7. I'm surprised that it's 45%. Might go down (?) with longer contex

I wanted to better understand the potential impact for the tokenizer change from 4.6 and 4.7. I'm surprised that it's 45%. Might go down (?) with longer context answers but still surprising. It can be more than 2x for small prompts.

pawelduda 6mo ago
Not very encouraging for longer use, especially that the longer the conversation, the higher the chance the agent will go off the rails