Grok inference glitch causes infinite repetition loops

This kind of repetition collapse is a known vulnerability in LLM inference, but seeing it raw in a major consumer product is a stark reminder of the fragility of these systems. The failure mode usually happens when the model's sampling gets stuck on a single high-probability token and fails to apply repetition penalties correctly. Infinite loops during reasoning suggest the model's internal thought traces are failing to converge or resolve complex prompts. While humorous to users, these infinite inference loops waste massive amounts of server compute and degrade the overall user experience.