tokenization system review

Gigatoken

SIMD and cache-heavy bulk BPE can make first-pass tokenization much faster on large corpora, including Qwen tokenizers on Apple Silicon. No current bottleneck was solved. Adoption requires exact token-ID parity and bounded peak memory on the real tokenizer and corpus.

Disposition
parked
Study class
tokenization system review
Reviewed
2026-09-23
Evidence
Tracked source record

What mattered

SIMD and cache-heavy bulk BPE can make first-pass tokenization much faster on large corpora, including Qwen tokenizers on Apple Silicon.

What changed locally

The implementation is retained as an optional offline data-preparation accelerator rather than a default tokenizer dependency.

Evidence limits

No current bottleneck was solved. Adoption requires exact token-ID parity and bounded peak memory on the real tokenizer and corpus.

Retained disposition

Revisit only for multi-gigabyte corpora or measured tokenization above ten percent of run time.

The action is retained as study context rather than an active backlog item. Any implementation, download, training run, or benchmark requires a fresh scoped question under the repository's experiment gate.

How to read this dossier

This canonical page separates the retained repository decision from the material that informed it. A source may describe an external claim, historical measurement, local observation, or planned procedure; those evidence classes are not interchangeable. Follow the provenance links before reusing a number or method.

Status and disposition describe what PostTrainLLM retained when this record was indexed. They are not a live product promise, a newly run benchmark, or permission to restart historical work. Unknown or unmeasured fields remain unknown rather than being treated as zero.

To reuse the record, first name the exact claim you need and trace it to the linked source. Then check whether the original environment, model revision, data split, evaluator, hardware, and budget match the proposed use. If they do not, treat the record as a hypothesis or design reference and run the smallest fresh comparison that can falsify it. Preserve negative outcomes and regressions beside any improvement; a local win on one slice does not silently become a general capability claim.

Source provenance

The normalized record comes from docs/studies/registry.json. The links below are the tracked evidence and explanatory sources preserved with the record.