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

    Neuronpedia

    INDEX
    Explanations

    phrases or words indicating a higher position or quality

    oai_token-act-pair · gpt-3.5-turbo

    references to quantities or levels that exceed a certain threshold

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Ring
    -0.73
    Mail
    -0.73
    Sport
    -0.69
     Dispatch
    -0.69
     CTR
    -0.68
    Access
    -0.66
     Centers
    -0.66
    Wheel
    -0.65
     Blades
    -0.65
    ck
    -0.64
    POSITIVE LOGITS
     above
    1.01
    above
    0.92
     Above
    0.87
    ceed
    0.86
    perty
    0.81
    hetically
    0.76
     mentioned
    0.76
    mentioned
    0.76
     foregoing
    0.75
     below
    0.74
    Act Density 0.016%

    No Known Activations