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

    Neuronpedia

    INDEX
    Explanations

    time durations in the format of "x and a half years"

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "half" or its variations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    cler
    -0.65
     accompanies
    -0.61
    ocratic
    -0.61
    orship
    -0.61
    rise
    -0.59
     preced
    -0.58
    cius
    -0.58
    ORTS
    -0.58
    achus
    -0.58
    Trend
    -0.58
    POSITIVE LOGITS
    half
    0.91
    heartedly
    0.85
    foundland
    0.82
    moon
    0.80
    quartered
    0.79
    pipe
    0.78
    eworks
    0.77
     thirds
    0.77
    bye
    0.71
    hearted
    0.70
    Act Density 0.004%

    No Known Activations