Hugging Face LLM Course: subword tokenization illustration Downloaded: 2026-10-23 (Asia/Bangkok) Author/source: Hugging Face / huggingface-course Lesson: https://huggingface.co/learn/llm-course/en/chapter2/4#subword-tokenization Original asset: https://huggingface.co/datasets/huggingface-course/documentation-images/resolve/96e71ad956904da670e6827274fc2c9b735b09d5/en/chapter2/bpe_subword.svg Dataset revision: 96e71ad956904da670e6827274fc2c9b735b09d5 License declaration: https://huggingface.co/datasets/huggingface-course/documentation-images/blob/96e71ad956904da670e6827274fc2c9b735b09d5/README.md License: Apache License 2.0, https://www.apache.org/licenses/LICENSE-2.0 Retained full license: /articles/credits/huggingface-course-Apache-2.0.txt Retained license declaration: /articles/credits/huggingface-course-images-96e71ad9-README.md Local asset: /articles/ai-token-basics/subword-tokenization.svg Original dimensions: 1826 x 172 Modifications: none; original SVG bytes retained. Responsive display uses horizontal scrolling at narrow widths. This is an upstream conceptual illustration, not a KrobClick model test or a universal segmentation rule.