jp-Lydia-325 is the latest in my attempt to learn TI's.<ul><li>Elle is not her real name.</li><li>I made this embedding with 12 images. These aren't photoshoot images just candid's with a variety of backgrounds which I tried to remove via the text files when it was made.</li><li>While this model is close it does not reproduce any of the training images exactly. I may do another version.</li><li>12 images trained for 325 steps.</li><li>12 vectors per token.</li><li>I used the following progressive learning rate 0.05:10, 0.02:20, 0.01:60, 0.005:200, 0.002:500, 0.001:3000, 0.0005</li><li>Batch size of 4 with 3 gradient accumulation steps.</li></ul>As always I appreciate your feedback and look forward to seeing what you all create.

Files with SHA256: 816ecd536f64019f3e8e0056d0e77ab54ac7b53a982576095610da205f75d6b0