41554 blogs · [ { "id": "01a0876d-c545-7253-aa6b-2bee530de579", "title": "Learning MegaGem, from self-play to price discovery", "url": "https://djdumpling.github.io/2026/08/09/megagem.html", "published_at": "2026-08-09T04:00:00+00:00" }, { "id": "01a0876d-c545-7253-aa6b-2bee532dfd78", "title": "hunting for headroom on modded-nanoGPT (WR #82)", "url": "https://djdumpling.github.io/2026/05/27/modded-nanoGPT-WR.html", "published_at": "2026-05-27T04:00:00+00:00" }, { "id": "01a0876d-c545-7253-aa6b-2bee5341fca2", "title": "frontier model training methodologies", "url": "https://djdumpling.github.io/2026/01/31/frontier_training.html", "published_at": "2026-01-31T05:00:00+00:00" }, { "id": "01a0876d-c545-7253-aa6b-2bee542517db", "title": "activation engineering for privacy protection in LLMs", "url": "https://djdumpling.github.io/2025/12/14/SEA-privacy.html", "published_at": "2025-12-14T05:00:00+00:00" }, { "id": "01a0876d-c545-7253-aa6b-2bee54a50f9f", "title": "combinatorial reasoning environments for LLMs and RL", "url": "https://djdumpling.github.io/2025/11/24/rl_envs.html", "published_at": "2025-11-24T05:00:00+00:00" }, { "id": "01a0876d-c545-7253-aa6b-2bee55146337", "title": "whirlwind of PPO and RLHF for LLMs from scratch", "url": "https://djdumpling.github.io/2025/08/04/rlhf_gpt2.html", "published_at": "2025-08-04T04:00:00+00:00" } ] posts Claim your blog

Blog · corpus.blog/blogs/djdumpling.github.io/posts