catid (e/acc) on Twitter: "## Can open-source LLMs detect bugs in C++ code? No: ``` LLaMa 65B (4-bit GPTQ) model: 1 false alarms in 15 good examples.  Detects 0 of 13 bugs. Baize 30B (8-bit) model: 0 false alarms in 15 good examples.  Detects 1 of 13 bugs. Galpaca 30B (8-bit) model: 0 false alarms in 15…" / Twitter

https://twitter.com/i/web/status/1647155706548695040