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

    Neuronpedia

    INDEX
    Explanations

    morphology

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Bart
    -0.07
     Bild
    -0.07
     anti
    -0.07
    Tu
    -0.07
     send
    -0.07
    (trans
    -0.07
     approx
    -0.06
    ']))↵
    -0.06
    搜索
    -0.06
    \↵
    -0.06
    POSITIVE LOGITS
     présence
    0.07
     twilight
    0.07
    跟我
    0.07
    LError
    0.07
    arrow
    0.06
     Fellowship
    0.06
    -working
    0.06
    稷
    0.06
    aklı
    0.06
    缡
    0.06
    Act Density 0.063%

    No Known Activations