Daniel Trejo

The project does not aim to train frontier-scale foundation models or compete in large-scale pretraining. Instead, it focuses on experimental inference-layer research involving: - residual trajectory observation, - embedding-space analysis, - continuity persistence testing, - lightweight modulation systems, - and inferential collapse dynamics inside transformer architectures.