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

    Neuronpedia

    INDEX
    Explanations

    nature

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     rivers
    -1.14
     Rivers
    -1.08
    Rivers
    -1.06
     streams
    -1.02
     creeks
    -0.94
    WebVitals
    -0.93
     Streams
    -0.91
    Streams
    -0.89
    rivers
    -0.88
    streams
    -0.86
    POSITIVE LOGITS
     and
    0.73
    .
    0.69
    ,
    0.68
    ides
    0.62
     among
    0.60
     in
    0.57
     composed
    0.51
     for
    0.51
     are
    0.49
     that
    0.47
    Act Density 0.083%

    No Known Activations