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

    Neuronpedia

    INDEX
    Explanations

    entities related to infrastructure and project planning, focusing on announcements and specifics such as costs and completion estimates

    oai_token-act-pair · gpt-3.5-turbo

    phrases that contain commas, indicating lists or separations of ideas

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    uit
    -0.63
    unic
    -0.60
    laughter
    -0.58
    agin
    -0.58
    iny
    -0.57
    jured
    -0.56
    lier
    -0.56
    bug
    -0.55
    mos
    -0.55
    uber
    -0.55
    POSITIVE LOGITS
     namely
    1.18
     albeit
    1.10
     although
    1.07
     regardless
    1.06
     irrespective
    1.04
     including
    1.00
     which
    0.99
     despite
    0.96
     however
    0.95
    which
    0.93
    Act Density 0.556%

    No Known Activations