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

    Neuronpedia

    INDEX
    Explanations

    text and image inputs

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     पैदा
    0.51
    也許
    0.47
    常に
    0.46
    ligence
    0.46
     इंद्र
    0.46
     पेर
    0.46
     መጨ
    0.46
    すべての
    0.45
     हमेशा
    0.45
     ಸ್ಥ
    0.45
    POSITIVE LOGITS
    i
    0.49
     ен
    0.47
     opaque
    0.46
    ЕЛ
    0.45
     carbon
    0.44
     macaroni
    0.44
     dipped
    0.43
     held
    0.43
     block
    0.42
     carbone
    0.42
    Act Density 0.002%

    No Known Activations