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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    овал
    -0.07
     improves
    -0.07
     relatives
    -0.07
    sst
    -0.07
     Witch
    -0.07
     Skull
    -0.06
    好き
    -0.06
    unes
    -0.06
     lifes
    -0.06
     additionally
    -0.06
    POSITIVE LOGITS
    Css
    0.07
    GroupName
    0.07
    זיכרון
    0.07
    European
    0.07
     pil
    0.07
    ざ
    0.06
    Contained
    0.06
    _variance
    0.06
    /,↵
    0.06
     ERA
    0.06
    Act Density 0.023%

    No Known Activations