# NOTICE — Neuratal NTokenizer v0.3-polished model This NOTICE covers the attribution and ShareAlike terms attached to the model artifact `ntokenizer-0.3-256k.ntok` and to the `tokenizer.json` export included in this repository. 1. The model is a machine-learned derivative of a 1.96 GB training corpus dominated by CC BY-SA/GFDL text from Wikimedia projects (simplewiki, trwiki, hiwiki, eswikisource, arwikisource, enwikibooks, jawikisource, kowikisource, zhwiktionary, jawiktionary, kowiktionary, trwiktionary and others). 2. It is therefore distributed under **CC BY-SA 4.0**, not Apache-2.0. ShareAlike applies to adaptations of the model. 3. Attribution pattern: `" contributors, , "`. Example: "Simple English Wikipedia contributors, CC BY-SA 4.0, https://simple.wikipedia.org". The complete per-source table (dumps, licenses, URLs, SHA-256) is in `data/v0.2/LICENSES.md` of the source repository () and summarized in this repo's `training_manifest.json` (fingerprints only; no corpus text included). 4. CC BY sources (arXiv per-article CC BY 4.0 — 187 papers; enwikinews CC BY 2.5/4.0; Tatoeba CC BY 2.0 FR) require attribution, no ShareAlike. 5. Public-domain sources (Project Gutenberg) and permissively licensed code (CPython PSF 2.0, TypeScript Apache-2.0, Go BSD-3-Clause, Rust MIT OR Apache-2.0) impose only notice retention, satisfied by this NOTICE. 6. No training corpus is distributed. No benchmark file is distributed. The benchmark used in evaluation is a development/validation set. 7. Founder and creator: Rüzgar Erdem. Neuratal is an independent AI research and software project; it is not claimed to be an incorporated company. The model is an independent implementation; no third-party tokenizer vocabulary, merge table, or implementation was used. 8. If you cannot comply with CC BY-SA 4.0, do not redistribute the model. The NTokenizer source code remains MIT OR Apache-2.0.