← Back to context

Comment by luckystarr

7 days ago

Tried it on Android and got "!!!!!!!!!!!!!" for answers.

The qwen models really seem to have this as a failure mode, its so annoying having a proper trace ending up in !!!!!! Garbage.

  • Oh, interesting. "!" is token id 0 in the Qwen tokenizer; I wonder if there's some tokenizer shenanigans either in inference or training that end up causing this specific behavior.

  • Wait in a regular sentence, what is the probability of "!!!" being followed by "!"?

    Sounds like the model is not following a proper probabilistic choice here, so maybe more a programming error than a model training error.

That's what happens when you quant too hard. I'm working on quant strats and evals for the same underlying qwen 27b models.

When I saw 27b on a phone, I thought not fitting, big phone, or aggressive quant. NVFP4 still takes 27G before KV cache.