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

    Neuronpedia

    INDEX
    Explanations

    BPA

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    eyond
    -0.09
    ishop
    -0.08
    ROWN
    -0.08
    acteria
    -0.07
    ENCH
    -0.07
    oulder
    -0.07
    ULK
    -0.07
    rescia
    -0.07
    acterial
    -0.06
    ulk
    -0.06
    POSITIVE LOGITS
     bfs
    0.08
     BDS
    0.07
    ordinal
    0.07
     literally
    0.07
    募
    0.06
     tua
    0.06
    들을
    0.06
    sizei
    0.06
     Essentially
    0.06
    なら
    0.06
    Act Density 0.016%

    No Known Activations