INDEX
    Explanations
    New Auto-Interp
    Negative Logits
    amping
    -0.09
    onni
    -0.08
     Reciprocity
    -0.08
    -0.08
     Scalar
    -0.08
     Paid
    -0.08
    .EM
    -0.08
     Budget
    -0.08
    ingtone
    -0.08
     Scalars
    -0.07
    POSITIVE LOGITS
     recount
    0.09
     narrator
    0.09
     narr
    0.08
     рассказ
    0.08
     anlat
    0.08
     describing
    0.08
     narration
    0.08
     imper
    0.08
    0.08
     beschreibt
    0.07
    Act Density 0.024%

    No Known Activations