jevVIDEO
Scoring Text DeepDream evolution with Jev
The author used Codex to build a single-file HTML page where 36 candidate sentences per generation are scored by Jev through a classifier, pushing "longing" from 56% to 93%.
Nicholas Dunzelman
@nicdunzArchived 2026.09.20Originally posted 2026.09.20
Nicholas Dunzelman @nicdunzOriginally posted 2026.09.20
i made Text DeepDream with Codex: random sentences evolve toward a concept. 36 candidates per generation, mutated locally. Jev scores them via @michael_chomsky’s classifier dot dev; the best survive. watch longing go from 56% to 93%. one HTML file.
x.com/nicdunz/status/2101363031028412534Archived 2026.09.20 10:13