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

    Neuronpedia

    INDEX
    Explanations

    phrases related to sources of power or energy

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to powering or enabling technologies

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    esan
    -0.81
    roma
    -0.74
     Bei
    -0.70
    ordan
    -0.69
    ei
    -0.69
    eal
    -0.68
    edi
    -0.67
     Debor
    -0.67
    ort
    -0.66
    innon
    -0.64
    POSITIVE LOGITS
     powered
    1.03
     Powered
    0.86
     powering
    0.86
     handc
    0.81
    powered
    0.80
     propelled
    0.79
    wagen
    0.76
    anche
    0.72
     outage
    0.72
    iences
    0.72
    Act Density 0.012%

    No Known Activations