INDEX
    Explanations

    The neuron detects occurrences of the word “wholesale.”

    New Auto-Interp
    Negative Logits
    běhu
    -0.07
    -0.07
    ΑΔ
    -0.06
    μμ
    -0.06
    §Ã
    -0.06
    pour
    -0.06
     Draws
    -0.06
    ічні
    -0.06
     Physical
    -0.06
    ivirus
    -0.06
    POSITIVE LOGITS
     wholesale
    0.12
    olesale
    0.08
     wholes
    0.07
    toDouble
    0.07
     annot
    0.07
     wholesalers
    0.07
     OEM
    0.07
    -le
    0.07
     Wholesale
    0.07
     institutional
    0.07
    Act Density 0.005%

    No Known Activations