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

    Neuronpedia

    INDEX
    Explanations

    study followed by findings

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    baş
    -0.94
     السياس
    -0.94
     transitioning
    -0.93
    ита
    -0.93
     külön
    -0.93
     memiliki
    -0.91
     JLabel
    -0.89
     jedynie
    -0.89
    варя
    -0.89
    ază
    -0.88
    POSITIVE LOGITS
     appeared
    0.92
     investigates
    0.90
     weitgehend
    0.90
     möglichst
    0.89
     investigate
    0.88
     affords
    0.87
     slop
    0.86
     occupies
    0.84
     coincide
    0.84
     regarded
    0.83
    Act Density 0.038%

    No Known Activations