INDEX
    Explanations

    The neuron is specialized to pick out version identifiers (especially hyphens and adjacent numerals) in dependency coordinates or file names.

    New Auto-Interp
    Negative Logits
    VV
    -0.07
     Kabul
    -0.06
     hvě
    -0.06
    Animate
    -0.06
     beraber
    -0.06
     Vector
    -0.06
     dagen
    -0.06
     Icon
    -0.06
     pregnant
    -0.06
    εξ
    -0.06
    POSITIVE LOGITS
     приб
    0.06
    _terminal
    0.06
    이다
    0.06
    ithmetic
    0.06
     nast
    0.06
    not
    0.06
    ↵↵↵↵↵↵↵↵↵↵↵↵↵↵
    0.06
     apparent
    0.06
    0.06
    شان
    0.06
    Act Density 0.013%

    No Known Activations