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

    Neuronpedia

    INDEX
    Explanations

    requesting content or instructions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     a
    0.99
     an
    0.84
    0.77
     in
    0.71
    ir
    0.70
    ti
    0.68
     Invisalign
    0.68
    k
    0.67
    j
    0.65
    sembles
    0.63
    POSITIVE LOGITS
    고
    0.73
    して
    0.67
    の
    0.67
    and
    0.66
    する
    0.66
    o
    0.65
    로
    0.64
    ми
    0.63
    ни
    0.62
    ウ
    0.62
    Act Density 1.798%

    No Known Activations