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

    Neuronpedia

    INDEX
    Explanations

    the word "unlike" and symbols that look like foreign quotes

    oai_token-act-pair · gemini-2.0-flash

    unlike

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     «
    -0.76
    unlike
    -0.73
     «<
    -0.72
     unlike
    -0.71
     («
    -0.68
     「
    -0.66
     getAccount
    -0.60
    ónica
    -0.60
     Receipt
    -0.59
    pital
    -0.57
    POSITIVE LOGITS
    <bos>
    2.63
    '
    0.75
    ’
    0.71
    The
    0.57
    reactstrap
    0.53
    ########.
    0.52
    <h2>
    0.52
     The
    0.51
    A
    0.50
     betweenstory
    0.48
    Act Density 0.326%

    No Known Activations