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

    Neuronpedia

    INDEX
    Explanations

    expressions commonly used for emphasis or strong emotion

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "Heck" and its variations, indicating a focus on expressions of exasperation or emphasis

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    vation
    -0.78
    uria
    -0.77
    urious
    -0.74
    ongyang
    -0.70
    xia
    -0.68
    ĻĤ
    -0.68
     NCT
    -0.65
     depression
    -0.61
     Spy
    -0.60
    secution
    -0.60
    POSITIVE LOGITS
    lers
    1.18
    enger
    0.99
    ling
    0.91
    bats
    0.91
    ler
    0.88
    led
    0.88
    linger
    0.86
    ington
    0.83
    ledged
    0.83
    ings
    0.82
    Act Density 0.036%

    No Known Activations