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

    Neuronpedia

    INDEX
    Explanations

    background

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     evasion
    -0.07
     unemployment
    -0.07
     preceding
    -0.06
     getaway
    -0.06
    lahoma
    -0.06
     hài
    -0.06
     supplementary
    -0.06
     Fla
    -0.06
     elsewhere
    -0.06
     St
    -0.06
    POSITIVE LOGITS
     thị
    0.07
    ัฒ
    0.07
    ίνα
    0.06
    755
    0.06
     Bombay
    0.06
     arab
    0.06
     asymmetric
    0.06
    ARS
    0.06
    dream
    0.06
     باق
    0.06
    Act Density 0.005%

    No Known Activations