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

    Neuronpedia

    INDEX
    Explanations

    origin and source of things

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     oleh
    0.99
     by
    0.85
    By
    0.84
    を行
    0.83
    .|
    0.82
     By
    0.81
     bylo
    0.79
     вед
    0.78
    が行
    0.77
     mà
    0.77
    POSITIVE LOGITS
     largement
    0.86
     largely
    0.80
    كز
    0.78
     mostly
    0.75
    antly
    0.75
    riamo
    0.72
     mainly
    0.72
     abbastanza
    0.70
    uates
    0.70
    mostly
    0.69
    Act Density 0.176%

    No Known Activations