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

    Neuronpedia

    INDEX
    Explanations

    bench

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    sure
    -0.07
    .InputStreamReader
    -0.07
    嘏
    -0.07
    Ў
    -0.07
    ляем
    -0.07
    jec
    -0.06
     Brend
    -0.06
    Sparse
    -0.06
     manslaughter
    -0.06
    楒
    -0.06
    POSITIVE LOGITS
     Bey
    0.07
    éd
    0.07
    蜜蜂
    0.07
     bots
    0.06
     produit
    0.06
     mex
    0.06
    ******/↵↵
    0.06
    inx
    0.06
     Dice
    0.06
    ?key
    0.06
    Act Density 0.002%

    No Known Activations