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

    Neuronpedia

    INDEX
    Explanations

    during

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Machines
    -0.06
    benchmark
    -0.06
     rewriting
    -0.06
     Coupon
    -0.06
    191
    -0.06
     Redemption
    -0.06
     svob
    -0.06
     земля
    -0.06
     สำ
    -0.06
     Formula
    -0.06
    POSITIVE LOGITS
    .vx
    0.07
     their
    0.06
    =q
    0.06
    .utf
    0.06
    'label
    0.06
    ολ
    0.06
    .top
    0.06
     и
    0.06
    _WM
    0.06
     hindsight
    0.06
    Act Density 0.014%

    No Known Activations