INDEX
    Explanations

    punctuation

    This neuron activates on tokens that are floating-point numbers (i.e. decimal numerical values).

    New Auto-Interp
    Negative Logits
     Bliss
    -0.07
     شرح
    -0.07
    Netflix
    -0.06
    copies
    -0.06
     Brill
    -0.06
    .getC
    -0.06
    markers
    -0.06
    itage
    -0.06
    ends
    -0.06
    _fake
    -0.06
    POSITIVE LOGITS
     метод
    0.07
     *}
    0.06
    _PIN
    0.06
    OrDefault
    0.06
    ニニ
    0.06
    ิภาพ
    0.06
    食品
    0.06
     ssize
    0.06
     runnable
    0.06
    ٫
    0.06
    Act Density 0.061%

    No Known Activations