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

    Neuronpedia

    INDEX
    Explanations

    the word "these."

    oai_token-act-pair · gpt-3.5-turbo

    the repeated phrase "these" with associated numerical values indicating importance or urgency

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    achus
    -0.85
     Pwr
    -0.79
    terness
    -0.77
    fecture
    -0.77
    ossier
    -0.77
    ixtape
    -0.73
    onis
    -0.73
    obook
    -0.73
    ppe
    -0.72
    zzle
    -0.71
    POSITIVE LOGITS
     kinds
    1.40
     sorts
    1.38
     types
    1.19
     shores
    1.09
     developments
    1.06
     situations
    1.03
     endeavors
    1.01
     occurrences
    1.01
     truths
    1.00
     facts
    1.00
    Act Density 0.111%

    No Known Activations