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

    Neuronpedia

    INDEX
    Explanations

    historical centuries

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    م
    1.28
    er
    1.19
    hand
    1.12
    لين
    1.08
    ার
    1.04
    nce
    1.01
    v
    1.00
    names
    0.98
    fourth
    0.98
    ie
    0.97
    POSITIVE LOGITS
     wenn
    1.02
     percentile
    0.97
    ीन
    0.94
     aus
    0.92
     ausz
    0.91
     się
    0.90
     onderdeel
    0.90
    ся
    0.89
     Fool
    0.87
     Ausgabe
    0.86
    Act Density 0.025%

    No Known Activations