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

    Neuronpedia

    INDEX
    Explanations

    the word "concrete", and sometimes "building", and "allow"

    oai_token-act-pair · gemini-2.0-flash

    concrete

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     cdti
    -1.04
     itſelf
    -1.02
     Efq
    -1.02
     myſelf
    -0.99
     iſt
    -0.99
     ―――――
    -0.98
     auffi
    -0.98
    Vidite
    -0.95
     tfsi
    -0.94
     crdi
    -0.93
    POSITIVE LOGITS
    !
    0.59
     .
    0.55
     I
    0.53
    </b>
    0.52
    /
    0.51
     дове
    0.50
     B
    0.50
    ia
    0.50
     Mo
    0.50
     επι
    0.50
    Act Density 2.444%

    No Known Activations