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

    Neuronpedia

    INDEX
    Explanations

    the word "full", and variations of the word "change"

    oai_token-act-pair · gemini-2.0-flash

    full or change

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     caufe
    -0.97
     ſeveral
    -0.93
     Theſe
    -0.92
     cauſe
    -0.92
     Jefus
    -0.91
     ſhall
    -0.89
     preſent
    -0.89
     juſ
    -0.88
     ſtand
    -0.88
     itſelf
    -0.88
    POSITIVE LOGITS
    <bos>
    0.64
      
    0.63
     and
    0.61
    ↵
    0.60
    IsContent
    0.58
    ,
    0.56
    ↵↵
    0.56
    hoeddwyd
    0.54
     I
    0.53
     for
    0.53
    Act Density 0.292%

    No Known Activations