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

    Neuronpedia

    INDEX
    Explanations

    Code and formatting

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     headed
    -0.07
     slaughter
    -0.07
     handheld
    -0.06
     turbo
    -0.06
     warped
    -0.06
     cathedral
    -0.06
     अव
    -0.06
     各
    -0.06
     manipulate
    -0.06
     Named
    -0.06
    POSITIVE LOGITS
    .socket
    0.08
    INE
    0.07
    券
    0.07
    _orient
    0.07
    ΄
    0.06
    ................
    0.06
    rame
    0.06
    ghi
    0.06
     기억
    0.06
    elerine
    0.06
    Act Density 0.006%

    No Known Activations