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

    Neuronpedia

    INDEX
    Explanations

    math and ratios

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     avoidance
    -0.08
     ACL
    -0.07
     WTO
    -0.07
    á
    -0.07
    🐀
    -0.07
    Germany
    -0.07
     населения
    -0.07
     scrim
    -0.06
    inho
    -0.06
    soever
    -0.06
    POSITIVE LOGITS
    undles
    0.08
     butterknife
    0.08
    (site
    0.08
    METHOD
    0.08
     hiatus
    0.08
    iable
    0.07
    CREASE
    0.07
    盷
    0.07
    .Scale
    0.07
    _TERM
    0.07
    Act Density 0.096%

    No Known Activations