INDEX
    Explanations

    The neuron consistently lights up on the names of lawyers and counsel listed in court‐case filings.

    New Auto-Interp
    Negative Logits
    logradouro
    -0.07
    .IGNORE
    -0.07
    _words
    -0.07
     звер
    -0.07
     sider
    -0.07
     داش
    -0.06
    QueryString
    -0.06
     oddly
    -0.06
    DIR
    -0.06
    <Token
    -0.06
    POSITIVE LOGITS
     whisk
    0.06
    ces
    0.06
    initialize
    0.06
     ACS
    0.06
     initialize
    0.06
    Legal
    0.06
    umi
    0.06
    ce
    0.06
    ubah
    0.06
     конферен
    0.06
    Act Density 0.006%

    No Known Activations