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

    Neuronpedia

    INDEX
    Explanations

    website URLs or technical references in the document

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

    stat.descriptive.test

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     noDo
    -0.84
     ***!
    -0.81
     EconPapers
    -0.75
    adaptiveStyles
    -0.71
    parsedMessage
    -0.65
    InjectAttribute
    -0.64
    enderror
    -0.62
     actionMode
    -0.61
    __':
    -0.61
     transfieras
    -0.60
    POSITIVE LOGITS
    Примітки
    0.41
    ByteBuf
    0.36
     instead
    0.36
    C
    0.36
     Instead
    0.36
    Che
    0.35
    ít
    0.35
    T
    0.35
     Schna
    0.35
    ly
    0.34
    Act Density 0.002%

    No Known Activations