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

    Neuronpedia

    INDEX
    Explanations

    FACT

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Integral
    -0.07
     Rif
    -0.07
    olah
    -0.06
    upy
    -0.06
    CLUS
    -0.06
     Officials
    -0.06
    迫
    -0.06
    bcm
    -0.06
     Levels
    -0.06
    lys
    -0.06
    POSITIVE LOGITS
    !↵↵↵
    0.07
    ETER
    0.07
    EXTERN
    0.07
    ....↵
    0.07
    ?↵↵↵
    0.07
    (directory
    0.07
    "↵↵
    0.06
    (o
    0.06
     failing
    0.06
    juries
    0.06
    Act Density 0.027%

    No Known Activations