jp-Julia2 is the latest in my attempt to learn TI's.<ul><li>I made this embedding with 8 stock images. Images were taken on a blank background .</li><li>While this model is close it does not reproduce any of the training images exactly. I may do another version.</li><li>8 images trained for 500 steps.</li><li>1 vector per token.</li><li>learning rate of .005</li><li>Batch size of 1 with 1 gradient accumulation steps.</li></ul>

Files with SHA256: 42c8fb798389accb4112d7f5bbaaba57571e1a1715975a8accefd6aa54d78f1f