A negative embedding focused on generating tonally cool images with added detail and realism.
NOTE: Since this focuses on generating cooler tones, you may have to push certain traits. As opposed to "redhead", use "((redhead))" if you are losing that aspect. This also tends to diminish smiles a little, so use weights or brackets to push them further if needed.
Only uses 5 tokens so as not to flood your negative token space.
Images were generated on my upcoming model.