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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating specific points or facts

    oai_token-act-pair · gpt-3.5-turbo

    the conjunction "that" and its repeated emphasis in sentences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    oses
    -0.81
    Tam
    -0.66
    cept
    -0.65
    zman
    -0.65
    Pont
    -0.65
    HO
    -0.65
    apsed
    -0.62
    agn
    -0.62
    MH
    -0.62
    eal
    -0.61
    POSITIVE LOGITS
     although
    0.94
     "[
    0.87
     there
    0.86
     whereas
    0.78
    chery
    0.78
     whilst
    0.77
     unlike
    0.76
     they
    0.72
     while
    0.71
     despite
    0.70
    Act Density 0.150%

    No Known Activations