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

    Neuronpedia

    INDEX
    Explanations

    numerical values associated with performance records

    oai_token-act-pair · gpt-3.5-turbo

    references to numerical performance metrics or statistics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    uana
    -0.82
    lez
    -0.76
    adies
    -0.76
    gypt
    -0.76
    opard
    -0.72
    ge
    -0.71
    AFTA
    -0.70
    xit
    -0.69
    pheus
    -0.67
    ita
    -0.67
    POSITIVE LOGITS
     breaker
    0.97
     record
    0.86
     Record
    0.81
     Breaker
    0.79
    record
    0.75
     shattering
    0.74
     records
    0.72
    LOCK
    0.72
     Records
    0.71
    breaking
    0.70
    Act Density 0.014%

    No Known Activations