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

    Neuronpedia

    INDEX
    Explanations

    what functions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     satisfactorily
    0.41
    ຢ່າງ
    0.41
     Matching
    0.41
    ję
    0.40
     expressed
    0.38
     कठोर
    0.38
     commensurate
    0.37
     Supporting
    0.37
     recoveries
    0.37
     supporting
    0.37
    POSITIVE LOGITS
    where
    0.51
     dónde
    0.47
     जहां
    0.45
    sendKeyDown
    0.45
     waar
    0.43
     όπου
    0.43
     where
    0.41
     där
    0.41
     hvor
    0.40
    Planck
    0.40
    Act Density 0.002%

    No Known Activations