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

    Neuronpedia

    INDEX
    Explanations

    Combinatorial math problems

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     learned
    -0.09
    {s
    -0.07
     который
    -0.07
     επ
    -0.07
     lernen
    -0.07
    lean
    -0.07
    ＿
    -0.07
     Atlanta
    -0.07
    �
    -0.07
     שבו
    -0.07
    POSITIVE LOGITS
     plast
    0.08
    Pinned
    0.08
     miscell
    0.08
     wrappers
    0.08
     Cpu
    0.08
    .Publish
    0.08
     particulares
    0.08
     제외
    0.08
     singled
    0.08
     Drivers
    0.08
    Act Density 0.105%

    No Known Activations