Repository navigation
Teach text-to-Token IDs through fixed BPE merges - #2
Merged
Merged
Conversation
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Adds the second reviewed AI learning scene: a bounded character-level BPE demonstration. A fixed English word expands into symbols, applies ranked adjacent merges, maps pieces to teaching IDs and decodes back to the same input. Switching rule versions isolates why character, word and Token counts differ.
Includes five words, source/spec, independent golden split/ID tests, span conservation, reversible decode and deterministic seeking. Learners predict a new word, receive intermediate-step feedback, then verify the result in the scene. The toy vocabulary, omitted byte/space/Chinese processing and lack of a real LLM are explicit.
Validation: npm run check passes including existing 721 planetary constraints. Desktop playback/completion, backward seeking, rule version 0↔3 and the fixed slider event-overwrite bug were checked; 390×844 layout and wrong/right quiz-to-diagram flow were inspected. No measured learner study or physical-device/screen-reader claim. Production record remains pending until the merged deployment is READY.