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

    Neuronpedia

    INDEX
    Explanations

    phrases with an emphasis on quantity or multitude

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    ħĭ
    -0.76
    eka
    -0.76
     Continued
    -0.70
    odan
    -0.68
    inth
    -0.64
    zinski
    -0.60
    arty
    -0.60
    Ħ¢
    -0.59
     Rover
    -0.57
    berus
    -0.56
    POSITIVE LOGITS
     crammed
    1.11
     vying
    0.92
     besides
    0.89
     imaginable
    0.87
     nowadays
    0.83
     Including
    0.78
     including
    0.78
     ranging
    0.75
     devoted
    0.75
     wasted
    0.73
    Act Density 0.477%

    No Known Activations