INDEX
    Explanations

    Not forcing/pressuring someone

    advice related to courting or romantic interactions.

    This neuron detects language about coercing or pressuring someone into doing something.

    New Auto-Interp
    Negative Logits
    -sensitive
    -0.06
    -0.06
     IID
    -0.06
    يير
    -0.06
    ertino
    -0.06
     Denn
    -0.06
    ürk
    -0.06
    оступ
    -0.05
     sorunu
    -0.05
    _typeDefinition
    -0.05
    POSITIVE LOGITS
    anzi
    0.08
    NR
    0.07
    .numberOf
    0.07
     wcs
    0.07
     cardinal
    0.06
    TRUE
    0.06
    0.06
    _cols
    0.06
    OUTPUT
    0.06
    ATURE
    0.06
    Act Density 0.015%

    No Known Activations