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

    Neuronpedia

    INDEX
    Explanations

    scientific articles

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     mop
    -0.08
    펶
    -0.08
    즌
    -0.07
    츠
    -0.07
     dame
    -0.07
    坥
    -0.06
    퓰
    -0.06
     ankle
    -0.06
    onents
    -0.06
    dT
    -0.06
    POSITIVE LOGITS
    Scalars
    0.08
     RID
    0.07
    discard
    0.07
    .mx
    0.07
     Wid
    0.07
    /password
    0.07
     metro
    0.07
     forgive
    0.07
    :@{
    0.07
     Said
    0.07
    Act Density 0.135%

    No Known Activations