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

    Neuronpedia

    INDEX
    Explanations

    references to ages expressed in the format "number-year-old"

    oai_token-act-pair · gpt-3.5-turbo

    references to age

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    achus
    -0.84
    undown
    -0.83
     plun
    -0.75
     subp
    -0.74
     lobb
    -0.73
    ktop
    -0.73
    hod
    -0.71
     cumbers
    -0.70
     tremend
    -0.68
    ickle
    -0.68
    POSITIVE LOGITS
     ago
    1.09
    long
    1.07
    ly
    0.87
    iversary
    0.86
    wide
    0.86
    lene
    0.84
    year
    0.80
     Ago
    0.78
    glass
    0.77
    len
    0.76
    Act Density 0.042%

    No Known Activations