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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating a sense of something being wrong or suspicious

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

    something is wrong or suspicious

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     anything
    -0.60
     Anything
    -0.49
    Anything
    -0.47
    anything
    -0.47
    exitRule
    -0.37
     ничего
    -0.37
     ANYTHING
    -0.36
    ByUserId
    -0.34
     cré
    -0.33
    usercontent
    -0.33
    POSITIVE LOGITS
     amiss
    0.52
     stimmt
    0.52
     fishy
    0.51
    ruptedException
    0.50
    jenost
    0.47
     ModelExpression
    0.47
    richTextPanel
    0.47
     misterioso
    0.47
    very
    0.46
     оригіналу
    0.46
    Act Density 0.018%

    No Known Activations