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

    Neuronpedia

    INDEX
    Explanations

    `.hs` file extensions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    tede
    1.05
    tri
    1.04
    taus
    0.98
    th
    0.93
    ătate
    0.93
    しさ
    0.93
    ikuti
    0.93
    ่
    0.93
    pooling
    0.92
    ted
    0.92
    POSITIVE LOGITS
     to
    1.48
    ني
    1.42
    (
    1.37
     be
    1.35
    ון
    1.30
    {
    1.25
    ל
    1.24
    بي
    1.21
     fünf
    1.10
     arbeit
    1.09
    Act Density 0.000%

    No Known Activations