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

    Neuronpedia

    INDEX
    Explanations

    trained and programmed

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     você
    0.60
     esté
    0.59
     اسرائی
    0.57
     esteja
    0.57
     rosé
    0.55
     enthalpy
    0.54
     portata
    0.54
     EMG
    0.53
     annotate
    0.52
     فوټبال
    0.51
    POSITIVE LOGITS
    which
    0.66
     upbringing
    0.58
    her
    0.57
    educated
    0.55
    chu
    0.53
    dirt
    0.53
    that
    0.52
    chh
    0.51
    观念
    0.51
     जीवन
    0.50
    Act Density 0.262%

    No Known Activations