Official code release for "SuperBPE: Space Travel for Language Models"
active 2025-04-10 → 2026-05-20 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 406 days from 2025-04-10 to 2026-05-20. Pushes: 14 total, peak 6 in a day. Pull requests: 2 total, peak 1 in a day. Issues: 14 total, peak 2 in a day. Comments: 24 total, peak 5 in a day. Stars: 35 total, peak 3 in a day.
- Pushes
- Pull requests
- Issues
- Comments
- Stars
Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.
Top contributors
Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity
| Contributor | Contributions | Pushes | PRs | Comments |
|---|---|---|---|---|
| PythonNut | 17 | 9 | 1 | 5 |
| alisawuffles | 11 | 5 | 0 | 5 |
| MattYoon | 4 | 0 | 0 | 2 |
| Iambestfeed | 4 | 0 | 0 | 4 |
| xiulinyang | 3 | 0 | 0 | 3 |
| yi-yvonne-yu | 2 | 0 | 0 | 0 |
| Yi-Yu-Yvonne | 2 | 0 | 0 | 1 |
| ShreyGanatra | 2 | 0 | 0 | 1 |
| narutatsuri | 2 | 0 | 0 | 1 |
| huyiwen | 1 | 0 | 0 | 1 |
| rrenaud | 1 | 0 | 0 | 0 |
| lsy641 | 1 | 0 | 0 | 0 |
| klasocki | 1 | 0 | 1 | 0 |
| wang99711123 | 1 | 0 | 0 | 0 |
| sjcheon86 | 1 | 0 | 0 | 1 |
Recent activity
Latest issues, pull requests and releases
- Issue comment#6PythonNut2026-05-03 07:28Require a huge amount of memory
- Issue comment#24ShreyGanatra2026-04-13 05:46pre-training olmo code missing
- Issue comment#23PythonNut2026-04-13 05:46SuperBPE produces garbage characters for non-English languages (Sinhala)
- Issue#24ShreyGanatra2026-04-01 04:32pre-training olmo code missing
- Issue comment#21PythonNut2026-01-17 21:48Is there a way to package this?
- Pull request#22klasocki2025-12-16 17:33
- Issue#20rrenaud2025-11-11 22:47Upstreaming minimal set of changes to HF tokenizers to get rid of the need for the fork
- Issue comment#19alisawuffles2025-11-04 23:22Stage 2 Ignores Outputs from Stage 1?
- Issue#19narutatsuri2025-10-31 02:24Stage 2 Ignores Outputs from Stage 1?
- Issue comment#14narutatsuri2025-10-31 02:07num_inherit_merges has no effect in Stage 2 tokenizer training
- Issue#17PythonNut2025-10-14 21:16Unusually slow training speed
- Issue#16PythonNut2025-10-14 21:10Please add a license
- Issue comment#15xiulinyang2025-10-06 03:29Incorporating superbpe into HF training pipeline
- Issue comment#15Iambestfeed2025-10-06 03:14Incorporating superbpe into HF training pipeline
- Issue comment#15xiulinyang2025-10-06 03:02Incorporating superbpe into HF training pipeline
- Issue comment#15Iambestfeed2025-10-06 02:25Incorporating superbpe into HF training pipeline
- Issue comment#15xiulinyang2025-10-06 02:02Incorporating superbpe into HF training pipeline
- Issue comment#17Iambestfeed2025-10-05 10:18Unusually slow training speed
- Issue comment#15Iambestfeed2025-10-05 06:31Incorporating superbpe into HF training pipeline
- Issue#14yi-yvonne-yu2025-09-08 03:05num_inherit_merges has no effect in Stage 2 tokenizer training
- Issue#13yi-yvonne-yu2025-09-06 13:40Request for tokenizer data-prep scripts (olmo-mix-p99-subset)
- Issue#12Yi-Yu-Yvonne2025-07-23 17:40extend_tokenizer result merges differ from original despite copying merges.txt
- Issue comment#7alisawuffles2025-07-07 18:27Citation request for a similar work at EMNLP 2023 paper
- Issue comment#4alisawuffles2025-07-01 00:06More comparison
- Issue comment#11alisawuffles2025-06-30 23:35Pre-Training data
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 35 stars here means stars gained during the window, not the repo's star count.