Weird output without the system
I got weird output without the system:
user:
hi
assistant:
,.. The question is about the number of ways to arrange the letters of the word "PROBLEM" such that no two vowels are adjacent. First, I need to identify the vowels and consonants in "PROBLEM". Let's spell it out: P, R, O, B, L, E, M. The vowels here are O ...
With the system("You are a helpful assistant. To answer the user's question, you first think about the reasoning process and then provide the user with the answer. The reasoning process and answer are enclosed within and tags, respectively, i.e., reasoning process here answer here ."):
user:
hi
assistant:
Okay, the user just said "hi". I should respond in a friendly and welcoming manner. Let me keep it simple and open-ended to encourage them to share more if they want. Maybe something like, "Hello! How can I assist you today?" That should work.
<answer> Hello! How can I assist you today? </answer>
The importance of system is out of my expectation, just leave the information for others who meet same situation.
Thanks for your attention!
We have included the system prompt in the tokenizer configuration, as it was used during both the SFT and RL stages. To ensure consistent output quality, we recommend including the same system prompt during actual usage; otherwise, the model's responses may be significantly affected.
Thanks for sharing the issue and thanks @howtain for explaining :)
We have updated the model card for the quantized version to better note this :)