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

    Neuronpedia

    INDEX
    Explanations

    statements indicating a transition or conclusion

    oai_token-act-pair · gpt-3.5-turbo

    the phrase "So" as a conversational opener or transition

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    exclusive
    -0.67
    ¢
    -0.63
    thro
    -0.62
     presentation
    -0.60
    ´
    -0.60
     ``(
    -0.59
    degree
    -0.58
    ASED
    -0.57
     Â
    -0.57
     Scand
    -0.57
    POSITIVE LOGITS
    oner
    1.25
    othes
    1.03
    bered
    1.03
    apy
    1.02
    fter
    0.98
    othe
    0.90
    oths
    0.89
    ooo
    0.87
    oooo
    0.86
    ften
    0.82
    Act Density 0.049%

    No Known Activations