41554 blogs · [ { "id": "01a0871f-8d91-7028-b7f1-21c200f5e963", "title": "LLM inference is nearly deterministic. We use this to audit providers", "url": "https://adamkarvonen.github.io/machine_learning/2025/11/28/difr.html", "published_at": "2025-11-28T19:46:58+00:00" }, { "id": "01a0871f-8d91-7028-b7f1-21c201ac9a19", "title": "Frontier AI Models Still Fail at Basic Physical Tasks: A Manufacturing Case Study", "url": "https://adamkarvonen.github.io/machine_learning/2025/04/13/llm-manufacturing-eval.html", "published_at": "2025-04-13T19:46:58+00:00" }, { "id": "01a0871f-8d91-7028-b7f1-21c201ea8326", "title": "Using an LLM perplexity filter to detect weight exfiltration", "url": "https://adamkarvonen.github.io/machine_learning/2024/07/21/weight-exfiltration.html", "published_at": "2024-07-21T19:46:58+00:00" }, { "id": "01a0871f-8d91-7028-b7f1-21c2026b01cd", "title": "Evaluating Sparse Autoencoders with Board Games", "url": "https://adamkarvonen.github.io/machine_learning/2024/06/12/sae-board-game-eval.html", "published_at": "2024-06-12T19:46:58+00:00" }, { "id": "01a0871f-8d91-7028-b7f1-21c202cafb8e", "title": "An Intuitive Explanation of Sparse Autoencoders for LLM Interpretability", "url": "https://adamkarvonen.github.io/machine_learning/2024/06/11/sae-intuitions.html", "published_at": "2024-06-11T19:46:58+00:00" }, { "id": "01a0871f-8d91-7028-b7f1-21c2033e2d20", "title": "Manipulating Chess-GPT’s World Model", "url": "https://adamkarvonen.github.io/machine_learning/2024/03/20/chess-gpt-interventions.html", "published_at": "2024-03-20T19:46:58+00:00" }, { "id": "01a0871f-8d91-7028-b7f1-21c2038ea312", "title": "Chess-GPT’s Internal World Model", "url": "https://adamkarvonen.github.io/machine_learning/2024/01/03/chess-world-models.html", "published_at": "2024-01-03T22:46:58+00:00" } ] posts Claim your blog

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