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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating that something is intentionally created or planned for a specific purpose

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate something has been intentionally created or crafted for a specific purpose

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    held
    -0.71
    dash
    -0.70
    Sharp
    -0.66
    rika
    -0.65
    oner
    -0.64
    Thom
    -0.64
    Ser
    -0.63
     Michaels
    -0.62
    enne
    -0.62
    Member
    -0.62
    POSITIVE LOGITS
    urally
    0.94
    ĸļ
    0.87
    ators
    0.86
     Parenthood
    0.85
    urat
    0.77
    irection
    0.76
    ifice
    0.75
    ating
    0.75
    emis
    0.73
    geries
    0.72
    Act Density 0.028%

    No Known Activations