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

    Neuronpedia

    INDEX
    Explanations

    instances of specific values or parameters in code

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "value" with varying numerical significance commonly found in programming contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ooks
    -0.77
    ocamp
    -0.76
    ility
    -0.71
    kefeller
    -0.70
    osaurus
    -0.70
    dar
    -0.69
    alyst
    -0.68
    udeb
    -0.68
    alm
    -0.67
     Citiz
    -0.65
    POSITIVE LOGITS
     value
    0.83
     values
    0.83
    values
    0.74
    EStream
    0.73
     Value
    0.69
    VALUE
    0.67
     Values
    0.65
    lda
    0.64
     Vance
    0.64
     proposition
    0.63
    Act Density 0.019%

    No Known Activations