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

    Neuronpedia

    INDEX
    Explanations

    instances of fraudulent or deceptive actions

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

    cheating, defrauding, robbing

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    TintMode
    -0.46
    OGND
    -0.44
     censiti
    -0.44
    AddTagHelper
    -0.44
    فایل‌لار
    -0.42
    gbaar
    -0.40
    RTEE
    -0.39
    smithy
    -0.39
     AssemblyCulture
    -0.38
    WillAppear
    -0.38
    POSITIVE LOGITS
     defraud
    0.63
     cheated
    0.59
     cheating
    0.56
     cheat
    0.56
     robbed
    0.55
    AndEndTag
    0.53
     defra
    0.52
    cheat
    0.52
     robbing
    0.51
     Cheat
    0.50
    Act Density 0.028%

    No Known Activations