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

    Neuronpedia

    INDEX
    Explanations

    numerical records or milestones

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "record" related to statistics or measurements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    adin
    -0.80
    creen
    -0.74
     Franch
    -0.74
    vette
    -0.72
    cation
    -0.69
    dot
    -0.67
    nery
    -0.67
    bite
    -0.66
    hiro
    -0.65
    cery
    -0.64
    POSITIVE LOGITS
     record
    1.07
     breaker
    0.97
    record
    0.96
     Record
    0.94
    breaking
    0.93
     holder
    0.88
     records
    0.85
     Breaker
    0.82
     Keeper
    0.81
     linkage
    0.80
    Act Density 0.023%

    No Known Activations