jp-j3a-1250 is the latest in my attempt to learn TI's.<ul><li>She is not a real person and does not reproduce any of the training images exactly, rather the look of a cute brunette. </li><li>22 images trained for 2250 steps.</li><li>12 vectors per token.</li><li>I used the following learning rate .004</li><li>Batch size of 1 with 1 gradient accumulation step.</li></ul>I'd appreciate any feedback you may have and would love to see what you come up with. Thanks and enjoy!

Files with SHA256: ae3874aa042eb47b0d1ba9937adede58beeb6ae5e9c693745445aaa43bb99793