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

    Neuronpedia

    INDEX
    Explanations

    news articles or reports featuring significant information or updates

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    rots
    -0.76
    ript
    -0.76
    amel
    -0.73
    phi
    -0.72
    atro
    -0.72
    ugi
    -0.71
    rot
    -0.70
    osal
    -0.69
    cel
    -0.66
    irie
    -0.66
    POSITIVE LOGITS
     ALSO
    1.00
     aloud
    1.00
    estone
    0.98
    soever
    0.95
    çīĪ
    0.89
    LESS
    0.84
    âĶģ
    0.80
    estones
    0.80
    istani
    0.77
     Writ
    0.76
    Act Density 0.336%

    No Known Activations