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

    Neuronpedia

    INDEX
    Explanations

    improvement

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    .ed
    -0.07
     opening
    -0.07
     MAC
    -0.07
    跨越
    -0.07
     which
    -0.07
     heavyweight
    -0.07
     Rim
    -0.07
     Бр
    -0.07
     mathematical
    -0.07
    -med
    -0.06
    POSITIVE LOGITS
     çağrı
    0.08
    זכור
    0.08
     yansı
    0.08
    unity
    0.07
    肸
    0.07
    mana
    0.07
     gw
    0.07
    便利
    0.07
    enty
    0.07
    tri
    0.07
    Act Density 0.263%

    No Known Activations