https://laojiahuo2003.github.io/https://laojiahuo2003.github.io/blog/2026-08-25-rmsnorm-notes/https://laojiahuo2003.github.io/blog/2026-08-26-gqa-notes/https://laojiahuo2003.github.io/blog/2026-08-26-rope-notes/https://laojiahuo2003.github.io/blog/2026-08-26-swiglu-notes/https://laojiahuo2003.github.io/blog/2026-09-25-ppo-dpo-grpo-notes/https://laojiahuo2003.github.io/blog/2026-09-26-agentic-rl-notes/https://laojiahuo2003.github.io/blog/2026-09-26-alignment-failures-notes/https://laojiahuo2003.github.io/blog/2026-09-26-flashattention-notes/https://laojiahuo2003.github.io/blog/2026-09-26-grpo-loss-notes/https://laojiahuo2003.github.io/blog/2026-09-26-grpo-variants-notes/https://laojiahuo2003.github.io/blog/2026-09-26-lora-notes/https://laojiahuo2003.github.io/blog/2026-09-26-loss-mask-notes/https://laojiahuo2003.github.io/blog/2026-09-26-parallelism-notes/https://laojiahuo2003.github.io/blog/2026-09-26-ppo-loss-gae-notes/https://laojiahuo2003.github.io/blog/2026-09-26-ppo-on-off-policy-notes/https://laojiahuo2003.github.io/blog/2026-09-26-quantization-granularity-notes/https://laojiahuo2003.github.io/blog/2026-09-26-sft-forgetting-notes/https://laojiahuo2003.github.io/blog/2026-09-26-training-memory-notes/https://laojiahuo2003.github.io/blog/2026-09-26-trust-region-ppo-notes/https://laojiahuo2003.github.io/blog/2026-09-27-llm-quantization-methods-notes/https://laojiahuo2003.github.io/blog/2026-09-27-llm-sampling-strategies-notes/https://laojiahuo2003.github.io/blog/2026-09-27-quantization-compute-cost-notes/https://laojiahuo2003.github.io/daily/https://laojiahuo2003.github.io/daily/2026-10-21/https://laojiahuo2003.github.io/daily/2026-10-22/https://laojiahuo2003.github.io/daily/2026-10-23/https://laojiahuo2003.github.io/daily/2026-10-24/https://laojiahuo2003.github.io/daily/2026-10-25/https://laojiahuo2003.github.io/daily/2026-10-26/https://laojiahuo2003.github.io/daily/2026-10-27/https://laojiahuo2003.github.io/daily/2026-10-28/https://laojiahuo2003.github.io/daily/2026-10-29/https://laojiahuo2003.github.io/daily/2026-10-30/https://laojiahuo2003.github.io/projects/https://laojiahuo2003.github.io/tags/%E5%90%8E%E8%AE%AD%E7%BB%83/https://laojiahuo2003.github.io/tags/%E5%A4%A7%E6%A8%A1%E5%9E%8B/https://laojiahuo2003.github.io/tags/%E5%BC%BA%E5%8C%96%E5%AD%A6%E4%B9%A0/https://laojiahuo2003.github.io/tags/%E6%A8%A1%E5%9E%8B%E7%BB%93%E6%9E%84/https://laojiahuo2003.github.io/tags/%E8%AE%AD%E7%BB%83%E5%B7%A5%E7%A8%8B/https://laojiahuo2003.github.io/weekly/