INDEX
    Explanations

    The neuron activates on author names in in-text citations.

    New Auto-Interp
    Negative Logits
    _private
    -0.07
    _CAM
    -0.06
    olar
    -0.06
    -water
    -0.06
    innie
    -0.06
    _FIFO
    -0.06
    ropri
    -0.06
    mic
    -0.06
    MBProgressHUD
    -0.06
    _CHANNEL
    -0.06
    POSITIVE LOGITS
    šť
    0.07
    Contents
    0.06
    &apos
    0.06
    ***↵
    0.06
    createElement
    0.06
     LEGO
    0.06
    ↵
    0.06
     stronghold
    0.06
     ordering
    0.06
    $
    0.06
    Act Density 0.014%

    No Known Activations