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

    Neuronpedia

    INDEX
    Explanations

    phrases related to lengths of time, specifically expressed in years

    oai_token-act-pair · gpt-3.5-turbo

    the phrase "years of" followed by a number

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    deck
    -0.72
    pload
    -0.71
    SourceFile
    -0.70
    heses
    -0.69
    uca
    -0.69
    eus
    -0.67
     Haram
    -0.67
     Cosponsors
    -0.66
    thren
    -0.66
    Reviewer
    -0.65
    POSITIVE LOGITS
     reckoning
    0.98
     probation
    0.87
     uninterrupted
    0.85
     struggle
    0.84
     frustration
    0.75
     elapsed
    0.73
     accumulated
    0.72
     imprisonment
    0.71
     silence
    0.71
     campaigning
    0.70
    Act Density 0.074%

    No Known Activations