sdprompt - lora to enhance __sdprompt__ wildcard trained with no data (animated gifs)

2000ステップで訓練済み:
art detailed concept painting digital highly sharp trending intricate focus illustration lighting portrait smooth elegant cinematic beautiful fantasy render realistic face style high hair body wearing eyes matte light character dramatic black details hyper dark ultra shot background detail epic volumetric girl photorealistic symmetrical cyberpunk blue masterpiece oil long 8k red female studio composition glowing hd vibrant quality artwork award man cute design fine dynamic shaded perfect painted resolution winning soft--

https://github.com/ntc-ai/conceptmod を使用

そして --save_opposite で逆に処理

最も頻出する単語（アーティスト名やサイト名を除く）でトークン上限を埋めるのが目的

/model/69651?modelVersionId=74311 と併用

アニメーション用のローラを -1 から 1 まで __sdprompt__ (/model/69651?modelVersionId=74311) を使って使用