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

    Neuronpedia

    INDEX
    Explanations

    structural

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    kode
    -0.07
    ُوا
    -0.07
     Tango
    -0.07
     Hund
    -0.06
     neue
    -0.06
     weg
    -0.06
    .bank
    -0.06
    (cid
    -0.06
    Nova
    -0.06
     babel
    -0.06
    POSITIVE LOGITS
    Previously
    0.07
    美
    0.07
    structural
    0.07
    国产
    0.07
    �
    0.06
     inflict
    0.06
    ，但
    0.06
    urances
    0.06
    QB
    0.06
    AMAGE
    0.06
    Act Density 0.004%

    No Known Activations