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

    Neuronpedia

    INDEX
    Explanations

    equations and code

    oai_token-act-pair · gemini-2.0-flash

    opening curly brackets

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    CharStream
    -0.66
    RAINT
    -0.64
    قایناق‌لار
    -0.55
    htdocs
    -0.55
    })->
    -0.54
    fcntl
    -0.51
    landt
    -0.51
    abestanden
    -0.50
    WebControls
    -0.49
    lihood
    -0.48
    POSITIVE LOGITS
     {
    1.20
     {
    0.93
    ,{
    0.91
     {{
    0.88
    {
    0.88
    {
    0.81
    ={{
    0.79
    ({
    0.74
     {'
    0.73
     {.
    0.72
    Act Density 2.706%

    No Known Activations