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

    Neuronpedia

    INDEX
    Explanations

    words related to large quantities or groups

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     Wiley
    -0.67
     Daisy
    -0.67
     Mistress
    -0.66
     Judges
    -0.64
    İĭ
    -0.63
     Apostles
    -0.60
    cale
    -0.60
     Roe
    -0.60
    ilk
    -0.60
     Misty
    -0.59
    POSITIVE LOGITS
    heads
    1.06
    head
    1.04
    ier
    0.86
    iness
    0.84
    iest
    0.84
    heading
    0.76
    haul
    0.76
    vertisement
    0.76
    ascript
    0.75
    weights
    0.74
    Act Density 0.040%

    No Known Activations