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

    Neuronpedia

    INDEX
    Explanations

    code delimiters and programming terms

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     merchandise
    0.86
     pixel
    0.83
     hydrochloric
    0.80
     moist
    0.80
     diode
    0.79
     merch
    0.79
     yummy
    0.79
     integration
    0.78
    出身
    0.78
     stole
    0.77
    POSITIVE LOGITS
     следует
    1.00
    é
    0.95
    á
    0.91
    ur
    0.91
     терпе
    0.89
     probabilidades
    0.85
    es
    0.84
    iend
    0.82
    een
    0.82
    aikum
    0.82
    Act Density 0.000%

    No Known Activations