INDEX
    Explanations

    oral administration or treatments

    This neuron detects the word "oral" specifically in medical or health-related contexts.

    New Auto-Interp
    Negative Logits
    -3.75
    {
    -2.91
    >
    -2.77
    ],
    -2.17
     “
    -2.17
     perfeitamente
    -2.16
     Which
    -2.14
     perfeita
    -2.09
    などは
    -2.05
     Provides
    -2.00
    POSITIVE LOGITS
     ハロウィン
    2.53
    قیمت
    2.44
    2.42
     invigor
    2.38
    2.36
     esteemed
    2.31
     supremely
    2.25
     multifaceted
    2.23
     飲
    2.22
     unparalleled
    2.20
    Act Density 0.012%

    No Known Activations