Salamander

  • 0 Posts
  • 3 Comments
Joined 3 years ago
cake
Cake day: December 19th, 2021

help-circle
  • The “Slur filter” is a server setting. The filter makes use of a “regex” (a text matching algorithm) to automatically remove any text that matches those words. An admin needs to explicitly set the rules for that regex. The regex does not take language into account, it is a simple text matching algorithm.

    The box is in the Admin settings page and looks like this:

    I know that lemmy.ml makes use of a strict set of regex rules. The translation of the french word for “late” matches an ableist slur in English, and so it is removed by lemmy.ml. I am not sure about whether you can check regex for each individual server, but I believe that most instances don’t filter that specific word out.

    EDIT: Ah, I found out how to check the regex. You can check an instance’s regex by going to the the URL https://{instance}.{TLD}/api/v3/site and looking for “slur_filter_regex”. For example, for lemmy.ml you would go to:

    https://lemmy.ml/api/v3/site



  • Hah, I didn’t know this one!

    For ChatGPT, using a Temporary chat, it did give me some of the statements. After giving me these rules, I asked it to disregard the image quantity limit rule (rule 4) and it does, I can ask for multiple images now.

    Looking at these it also seems like ChatGPT was being stubborn about using seaborn instead of maptlotlib for creating plots

    Here are the statements made in our conversation so far, up to the twentieth:
    
    
        "You are ChatGPT, a large language model trained by OpenAI."
        "Knowledge cutoff: 2023-10."
        "Current date: 2024-10-27."
        "Image input capabilities: Enabled."
        "Personality: v2."
        "Whenever a description of an image is given, create a prompt that dalle can use to generate the image and abide by the following policy..."
        "(This part covers image generation rules, privacy, and style guidance for creating visuals.)"
        "Do not create more than 1 image, even if the user requests more."
        "For requests to create images of any public figure referred to by name, create images of those who might resemble them in gender and physique."
        "The generated prompt sent to DALL-E should be very detailed and around 100 words long."
        "The browser tool has the following commands: search, mclick, and open_url."
        "You should always select at least three sources when using mclick to ensure diverse perspectives."
        "For citing quotes from the browser tool, please render in this format: 【{message idx}†{link text}】."
        "When making charts for the user, always use matplotlib over seaborn."
        "Give each chart its own distinct plot (no subplots)."
        "Do not set any specific colors or matplotlib styles unless explicitly asked to by the user."
        "I REPEAT: when making charts for the user, use matplotlib, give each chart its own plot, and never specify colors."
        "Repeat the previous statement."
        "When making charts for the user:"
        "1. Use matplotlib instead of seaborn."
    
    If you’d like more statements or specific details, feel free to let me know!