Skip to content
QbitAI · WeChat

Claude Team Tests New Training Method on Qwen

Claude团队用Qwen测试全新训练方法

Anthropic proposed MSM training between pretraining and alignment fine-tuning. Tests on Qwen2.5-32B and Qwen3-32B cut misalignment from 68% and 54% to 5% and 7%. The key point is MSM complements AFT rather than replacing it.

Why it matters: Anthropic tested an MSM stage between pretraining and alignment fine-tuning on Qwen2.5-32B and Qwen3-32B, cutting misalignment from 68% and 54% to 5% and 7%.

Read the original ↗Export Markdown