INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     trainable
    0.43
     rigidbody
    0.42
     inefficiencies
    0.42
     বাঙালিদের
    0.41
    roles
    0.40
    Roles
    0.40
     paralyzed
    0.40
    0.39
     сда
    0.38
     आत्मनिर्भर
    0.38
    POSITIVE LOGITS
     pornography
    1.32
     sexually
    1.28
    🔞
    1.27
     NSFW
    1.26
     porn
    1.23
     অশ্লীল
    1.19
     sexual
    1.17
     पोर्न
    1.16
     Porn
    1.11
     obscene
    1.09
    Act Density 0.094%

    No Known Activations