54,302 blogs · [ { "id": "01a087e5-ce52-73c8-8260-ff193f37d2fb", "title": "Hiding a prompt in a tokenizer", "url": "https://stephantul.github.io/blog/prompt/", "published_at": "2026-08-23T00:00:00+00:00" }, { "id": "01a087e5-ce52-73c8-8260-ff193ff08ef6", "title": "From continuing subword markers to word-initial markers", "url": "https://stephantul.github.io/blog/tnf/", "published_at": "2026-08-17T00:00:00+00:00" }, { "id": "01a087e5-ce52-73c8-8260-ff193fffc3d3", "title": "Skeletoken 0.4.0 release", "url": "https://stephantul.github.io/blog/skeletoken-update/", "published_at": "2026-08-02T00:00:00+00:00" }, { "id": "01a087e5-ce52-73c8-8260-ff1940dd7c8e", "title": "From Chesterton's fence to Chesterton's gap", "url": "https://stephantul.github.io/blog/unfence/", "published_at": "2026-06-14T00:00:00+00:00" }, { "id": "01a087e5-ce52-73c8-8260-ff194178a63b", "title": "Make something for someone", "url": "https://stephantul.github.io/blog/make-something/", "published_at": "2026-05-19T00:00:00+00:00" }, { "id": "01a087e5-ce52-73c8-8260-ff1941ac33dc", "title": "Scikit-learn's fit transform paradigm is probably not for you", "url": "https://stephantul.github.io/blog/fit-transform/", "published_at": "2026-05-17T00:00:00+00:00" }, { "id": "01a087e5-ce52-73c8-8260-ff1941bff89f", "title": "Evaluating static models on RTEB", "url": "https://stephantul.github.io/blog/rteb-static/", "published_at": "2025-10-08T00:00:00+00:00" }, { "id": "01a087e5-ce52-73c8-8260-ff19429ff293", "title": "Comparing PCA and MRL for static models", "url": "https://stephantul.github.io/blog/mrl-pca/", "published_at": "2025-10-06T00:00:00+00:00" }, { "id": "01a087e5-ce52-73c8-8260-ff19439e0581", "title": "Static late interaction models", "url": "https://stephantul.github.io/blog/static-colbert/", "published_at": "2025-09-30T00:00:00+00:00" }, { "id": "01a087e5-ce52-73c8-8260-ff19445ad590", "title": "Better Greedy Tokenizers: Handling WordPiece's [UNK] Problem", "url": "https://stephantul.github.io/blog/better-greedy/", "published_at": "2025-09-18T00:00:00+00:00" } ] posts Claim your blog

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