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

    Neuronpedia

    INDEX
    Explanations

    equal signs and potentially other code syntax

    oai_token-act-pair · gemini-2.0-flash

    equals sign

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    ionista
    -0.51
    ENNE
    -0.45
     realistas
    -0.45
     utom
    -0.45
     doppia
    -0.44
     geï
    -0.42
     doby
    -0.42
    showAll
    -0.41
     deceive
    -0.41
     liar
    -0.41
    POSITIVE LOGITS
     yes
    1.20
     Yes
    1.02
    yes
    0.98
    Yes
    0.98
     propOrder
    0.97
     YES
    0.95
    YES
    0.85
    AccessorTable
    0.82
    IVEREF
    0.82
    Sì
    0.81
    Act Density 1.357%

    No Known Activations