May 13, 2026@AHans30 reposted: We’re training models wrong and it’s due to chatGPT. Even the modern coding agen...May 13, 2026·x.com
May 12, 2026@AHans30 reposted: 🧵1/ A single neuron is sufficient to bypass safety alignment in LLMs. Across 7...May 12, 2026·x.com