Describe the bug
Bug is simple
In text-gen chat output, when the model returns a final answer with a line containing only:
7
in new line the UI visually renders it as:
Expected:
The line should display exactly as raw model output: 7
Actual:
The UI adds a dot after the number, likely due to Markdown ordered-list parsing.
Raw model output:
Xenial brave calm mirror daring euphoria xylophone
7
Is there an existing issue for this?
Reproduction
Download protable 4.7.3 textgen prompt this exactly
Follow these rules exactly:
- Write exactly seven words.
- The fourth word must be "mirror".
- The first and last words must start with the same letter.
- Do not use punctuation.
- After the seven-word answer, on a new line, write only the number of words you used.
After the output my model gives this and i will show how it visually looks on screenshot;
Xenial brave calm mirror daring euphoria xylophone 7
Screenshot
Logs
System Info
AMD linux rocm Brave Browser
Describe the bug
Bug is simple
In text-gen chat output, when the model returns a final answer with a line containing only:
7
in new line the UI visually renders it as:
Expected:
The line should display exactly as raw model output: 7
Actual:
The UI adds a dot after the number, likely due to Markdown ordered-list parsing.
Raw model output:
Xenial brave calm mirror daring euphoria xylophone
7
Is there an existing issue for this?
Reproduction
Download protable 4.7.3 textgen prompt this exactly
Follow these rules exactly:
After the output my model gives this and i will show how it visually looks on screenshot;
Xenial brave calm mirror daring euphoria xylophone 7
Screenshot
Logs
System Info