INDEX
    Explanations

    online conversations

    The main thing this neuron does is pick up on vague, non-specific qualifiers and approximations (e.g. “some,” “kind of,” “something,” “rough estimate,” “timeline”).

    New Auto-Interp
    Negative Logits
    IAN
    -0.07
    OG
    -0.06
     rated
    -0.06
    (serializer
    -0.06
    felt
    -0.06
     SEL
    -0.06
    ادگی
    -0.06
    ชาย
    -0.06
     pack
    -0.06
    放送
    -0.06
    POSITIVE LOGITS
     Πλη
    0.07
    .decor
    0.07
     spd
    0.07
    .FindElement
    0.07
     '<
    0.06
    िए
    0.06
    _Db
    0.06
     сервер
    0.06
    .Alignment
    0.06
     Aspect
    0.06
    Act Density 0.082%

    No Known Activations