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

    Neuronpedia

    INDEX
    Explanations

    say "ratio"

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    (year
    -0.08
     Sized
    -0.07
    .price
    -0.07
     кого
    -0.07
    <N
    -0.07
    Bearer
    -0.07
    肩
    -0.07
    (code
    -0.07
    Miami
    -0.07
     Forward
    -0.07
    POSITIVE LOGITS
    ucch
    0.06
     Rt
    0.06
    esper
    0.06
     Stateless
    0.06
    owo
    0.06
    meyen
    0.06
    ิจารณ
    0.06
     yapmak
    0.06
     centerX
    0.06
    gue
    0.06
    Act Density 0.017%

    No Known Activations