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

    Neuronpedia

    INDEX
    Explanations

    code and explanations

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Ꮜ
    0.65
    Their
    0.64
    Swatch
    0.63
    מצע
    0.62
    scrire
    0.61
    Seeing
    0.60
    rectionType
    0.60
     presentan
    0.59
    clipped
    0.59
    Introducing
    0.59
    POSITIVE LOGITS
     Preparation
    0.65
     preparation
    0.63
     Prepar
    0.59
     priprav
    0.59
     language
    0.58
     Materials
    0.58
     준비
    0.58
     Jesuit
    0.57
     Rao
    0.56
     selected
    0.55
    Act Density 0.002%

    No Known Activations