INDEX
    Explanations

    This neuron detects chemical ion notations, especially sodium (Na⁺), potassium (K⁺), and related ionic species names.

    New Auto-Interp
    Negative Logits
    इन
    -0.07
     scraper
    -0.07
     twe
    -0.07
    werp
    -0.06
     Icons
    -0.06
    .tem
    -0.06
    intestinal
    -0.06
     خوش
    -0.06
     artic
    -0.06
    xbc
    -0.06
    POSITIVE LOGITS
     dating
    0.07
     risks
    0.07
    нима
    0.06
    Decoder
    0.06
    unsqueeze
    0.06
     mix
    0.06
    !'↵
    0.06
     cooker
    0.06
    她们
    0.06
     акт
    0.06
    Act Density 0.006%

    No Known Activations