Correct the card: the model is now trained on the 19.2 MB TinyStories validation set (the actual #17 request). Fixes param count (100,936 unique; the earlier 106,568 double-counted the tied head) and reports the honest memorization result (ratio 1.02x, verified reproducible).
Fix false 'fulfils request #17' claim: the model was trained on 191 MB web text, not the requested TinyStories-validation-only experiment. Published as a from-scratch demonstration, not the requested experiment.
