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

    Neuronpedia

    INDEX
    Explanations

    over

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     shareholder
    -0.07
    und
    -0.06
     diagrams
    -0.06
    nex
    -0.06
     certain
    -0.06
    Dialogue
    -0.06
    rál
    -0.06
     совсем
    -0.06
    -ca
    -0.06
    Disc
    -0.06
    POSITIVE LOGITS
     Bag
    0.07
     قالب
    0.07
     İslâm
    0.07
     quantidade
    0.06
     quotient
    0.06
     UserRole
    0.06
     스타
    0.06
     paylaş
    0.06
    icaid
    0.06
    .Elapsed
    0.06
    Act Density 0.017%

    No Known Activations