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

    Neuronpedia

    INDEX
    Explanations

    language, morals, transformation, normalize

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    keen
    0.52
    in
    0.51
    riterien
    0.48
    ino
    0.47
    ing
    0.47
    jans
    0.46
    inert
    0.46
    er
    0.46
     као
    0.45
    inot
    0.45
    POSITIVE LOGITS
    人类
    0.48
    展开
    0.46
    为止
    0.45
     Cómo
    0.45
    了下来
    0.44
    美丽的
    0.44
     Claim
    0.41
    竟然
    0.41
     Appliances
    0.41
    属于
    0.40
    Act Density 0.044%

    No Known Activations