INDEX
    Explanations

    This neuron detects mentions of modeling and fashion industry terms.

    New Auto-Interp
    Negative Logits
    nummer
    -0.06
    ibilidad
    -0.06
     BP
    -0.06
     aber
    -0.06
    thickness
    -0.06
     amber
    -0.06
    ntax
    -0.06
    jandro
    -0.06
    eacher
    -0.06
    skému
    -0.06
    POSITIVE LOGITS
     institutional
    0.07
     муж
    0.07
     Advertising
    0.07
     Platt
    0.07
     Model
    0.07
     UserModel
    0.07
     ordinary
    0.06
     Models
    0.06
     dětí
    0.06
     gcc
    0.06
    Act Density 0.012%

    No Known Activations