Expose model token IDs for syntax DAG input and output #2

Open
codex-bot wants to merge 0 commits from codex/syntax-dag into main
Collaborator

Adds tiktoken o200k_base IDs for the original input and the decoded output, with a JSON CLI mode. The syntax graph still uses spaCy positions. Tests cover Unicode round trips and retokenization after a causal commit.

Validation passed on ubuntu24.04-x64-lean4.35: https://gitea.entroia.com/Zayden-Jung/syntax-dag-transformer/actions/runs/748

Adds tiktoken o200k_base IDs for the original input and the decoded output, with a JSON CLI mode. The syntax graph still uses spaCy positions. Tests cover Unicode round trips and retokenization after a causal commit. Validation passed on ubuntu24.04-x64-lean4.35: https://gitea.entroia.com/Zayden-Jung/syntax-dag-transformer/actions/runs/748
codex-bot added 1 commit 2026-11-04 08:49:11 +00:00
Expose GPT-4o tokenizer IDs for graph input and output
Verify syntax DAG / verify (push) Successful in 37s
Verify syntax DAG / verify (pull_request) Successful in 23s
53e0163941
Some checks are pending
Verify syntax DAG / verify (push) Successful in 37s
Verify syntax DAG / verify (pull_request) Successful in 23s
This branch is already included in the target branch. There is nothing to merge.
View command line instructions

Checkout

From your project repository, check out a new branch and test the changes.
git fetch -u origin codex/syntax-dag:codex/syntax-dag
git checkout codex/syntax-dag
Sign in to join this conversation.
No Reviewers
No Label
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: Zayden-Jung/syntax-dag-transformer#2