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

    Neuronpedia

    INDEX
    Explanations

    stones and pebbles

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ten
    0.82
    kannten
    0.80
    ু
    0.76
    O
    0.75
    Ketika
    0.74
    uigen
    0.73
    that
    0.73
    ।'
    0.71
    daten
    0.71
    tinger
    0.70
    POSITIVE LOGITS
    ل
    1.03
     pebble
    0.88
    ส
    0.88
     pebbles
    0.86
    л
    0.86
     stone
    0.85
    h
    0.82
     by
    0.80
    石
    0.79
     stones
    0.78
    Act Density 0.008%

    No Known Activations