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

    Neuronpedia

    INDEX
    Explanations

    code data and text

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     algum
    0.38
     onus
    0.36
     ello
    0.36
    瑄
    0.35
     Hombre
    0.35
     estuvo
    0.35
     correto
    0.35
     Implications
    0.35
     каких
    0.34
     якого
    0.34
    POSITIVE LOGITS
     sportive
    0.38
    fier
    0.37
     чёр
    0.37
    pleClass
    0.36
     duż
    0.36
    cią
    0.35
     svart
    0.35
     nightly
    0.35
     schwarz
    0.35
     カスタ
    0.34
    Act Density 0.031%

    No Known Activations