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

    Neuronpedia

    INDEX
    Explanations

    a followed by descriptor

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    '
    0.23
    ↵
    0.19
     (
    0.18
     (=
    0.18
     ,
    0.18
     [(
    0.17
    を使用
    0.17
     isotherms
    0.16
     which
    0.16
     when
    0.16
    POSITIVE LOGITS
     realist
    0.18
    רי
    0.18
     pragmatic
    0.15
    <unused277>
    0.15
     multitude
    0.15
    жли
    0.15
    ﻔ
    0.15
     confronto
    0.14
    ﻛ
    0.14
    хи
    0.14
    Act Density 1.602%

    No Known Activations