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

    Neuronpedia

    INDEX
    Explanations

    open a new thing

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    al
    0.69
    ir
    0.65
    y
    0.64
    ু
    0.64
    oon
    0.64
    in
    0.62
    ین
    0.62
    ל
    0.60
    ט
    0.60
    on
    0.59
    POSITIVE LOGITS
     ouvert
    1.09
     Opens
    1.06
     abiertos
    1.04
     Opening
    1.02
     aberta
    1.02
     opens
    1.01
     ouverte
    1.00
     öpp
    1.00
     Opened
    1.00
     åp
    0.99
    Act Density 0.149%

    No Known Activations