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

    Neuronpedia

    INDEX
    Explanations

    high [adjective]

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     веществ
    0.97
    חנו
    0.96
     logam
    0.92
     никакой
    0.88
     nyeri
    0.83
    יש
    0.82
    Ә
    0.82
    ंपा
    0.81
     heavier
    0.81
    적으로
    0.81
    POSITIVE LOGITS
     altitude
    1.34
    lighting
    1.31
    altitude
    1.30
     altitudes
    1.25
     kaliteli
    1.24
     priced
    1.23
     octane
    1.18
    idelity
    1.18
    erval
    1.14
    regarded
    1.13
    Act Density 0.238%

    No Known Activations