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

    Neuronpedia

    INDEX
    Explanations

    one-word phrases

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     diverso
    0.73
     successive
    0.73
     separate
    0.67
    separate
    0.64
    picable
    0.64
     fireworks
    0.64
     tien
    0.63
     brillo
    0.62
     Pues
    0.62
     forskellige
    0.62
    POSITIVE LOGITS
     liners
    1.20
    click
    1.17
     offs
    1.16
    offs
    1.11
    liners
    1.11
    liner
    1.10
    off
    1.09
     liner
    1.09
    Shot
    1.07
    sided
    1.06
    Act Density 0.070%

    No Known Activations