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

    Neuronpedia

    INDEX
    Explanations

    compound words or phrases containing the hyphen symbol ("-")

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     Pok
    -0.73
    Ń·
    -0.71
     Tycoon
    -0.67
     Miko
    -0.65
     charms
    -0.64
     Conquer
    -0.62
     Phi
    -0.62
     chants
    -0.60
     Zar
    -0.59
     preced
    -0.59
    POSITIVE LOGITS
    fledged
    1.51
    blown
    1.29
    sized
    1.19
    scale
    1.19
    bodied
    1.11
    length
    1.09
    contact
    1.04
    priced
    1.03
    blooded
    1.03
    thro
    1.03
    Act Density 0.020%

    No Known Activations