Notes

• Aug 22, 2026 Price the Reward Function — I measured what a code-execution reward costs a GRPO training loop in GPU-seconds. The median service time turns out to be the wrong number to look at.
• Aug 19, 2026 Map the Failure Boundary — I froze a Go1 locomotion policy and measured where it fails across 6,400 combinations of floor friction and lateral push, then retrained on the two gaps the map exposed and measured again.
• May 22, 2026 Monte Found a Decorative Channel and a Reward Exploit in One of Our Benchmarks — Monte's adversarial training exposed two failure modes in a MARL benchmark: a communication channel that looked load-bearing but wasn't, and a reward function that incentivized trivial policies.
• May 6, 2026 Where V-JEPA 2.1's Dense Features Hold Up (and Where They Don't) — A pre-registered V-JEPA 2.1 robustness study across all four model sizes, with deployment lessons.
• Jan 16, 2026 10,924x: The Instability Bomb at 1.7B Scale — Part 2 of the mHC reproduction series. I scaled from 10M to 1.7B parameters and watched Hyper-Connections hit 10,924x signal amplification. That's not a typo.
• Jan 11, 2026 DeepSeek's mHC: When Residual Connections Explode — DeepSeek replaced standard residuals with massive Hyper-Connections, and then watched them explode. I reproduced the 9x signal amplification to understand their fix: the Sinkhorn-Knopp algorithm. Part 1 of 2.