Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    strong emotional language and criticism directed towards people or groups

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    hyperbolic or harsh language

    np_acts-logits-general · gemini-2.5-flash-lite

    discussion about rhetoric and tone—especially references to strong, heated, biased, abusive, or hyperbolic language and how it’s being used or judged.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Negative Logits
     mouseY
    -0.50
     compromise
    -0.47
     Compromise
    -0.42
    IUrlHelper
    -0.42
     introducido
    -0.41
    Carriera
    -0.41
     '\\;'
    -0.40
    twimg
    -0.40
     bağlantılar
    -0.40
    日閲覧
    -0.40
    POSITIVE LOGITS
    WriteTagHelper
    0.42
     pinulongan
    0.40
     irrever
    0.39
     sözler
    0.38
     ویکی‌پدی
    0.38
    statusText
    0.35
     aggressive
    0.35
    attitude
    0.35
     оригіналу
    0.35
    emed
    0.34
    Act Density 0.077%

    No Known Activations