Jonnester's picture
Document corpus-v2 validation checks and cleaning results
ba7da9d verified
|
Raw History Blame Contribute Delete
10.9 kB
---
language:
- en
task_categories:
- text-generation
tags:
- tokenized
- pretraining
- corpus-v2
pretty_name: Purrence Hyperbullet 100B Corpus v2
---
# Purrence Hyperbullet 100B Corpus v2
This dataset provides 100,000,000,000 training input positions and 10,000,000 held-out validation positions. It is a custom memory-mapped token corpus, not a Parquet dataset for datasets.load_dataset. The exact build is identified by corpus-v2/manifest.json, corpus-v2/verification.json, and the UPLOAD_COMPLETE.json receipt. That receipt's data_revision pins the immutable commit whose new corpus files were verified.
The frozen tokenizer has SHA-256 fc240b741112488b639a711f502305601e3b8a1b01c5ae93a48b9ae55692dd33, vocabulary size 2,048, and structural document-boundary ID 2,047. The tokenizer JSON and its review/provenance artifacts are under corpus-v2/tokenizer/.
## Download and load
Download the corpus-v2/ prefix at the data_revision in UPLOAD_COMPLETE.json; preserve the prefix in the local directory. See corpus-v2/LOADER.md for a pinned download example, file formats, and the project's reproducible TokenLoader entry point. Training should use the manifest and source indexes; the physical token file contains stored positions that are not all selected into a split.
## Source composition
| Component | Upstream dataset | Pinned revision | Selector | Training positions | Validation positions |
| --- | --- | --- | --- | ---: | ---: |
| ufw_multi | openbmb/Ultra-FineWeb-L3 | bc3b1ba986fcaef6871b9790a413b16267c2de0f | data/ultrafineweb_en_l3/multi_style/ | 40,000,000,000 | 4,000,000 |
| ufw_qa | openbmb/Ultra-FineWeb-L3 | bc3b1ba986fcaef6871b9790a413b16267c2de0f | data/ultrafineweb_en_l3/qa/ | 20,000,000,000 | 2,000,000 |
| fineweb_mask | DATA-MASK/FineWeb-Mask | 8f5ab01fd5b74e47579961fdeae8a81c2bc5dca2 | chunk | 10,000,000,000 | 1,000,000 |
| math_conversation | openbmb/UltraData-Math | fe10db8efd35597fd7fcff8ff576b5ec4ea5ff87 | data/UltraData-Math-L3/Conversation-Synthetic/ | 3,750,000,000 | 375,000 |
| math_multi | openbmb/UltraData-Math | fe10db8efd35597fd7fcff8ff576b5ec4ea5ff87 | data/UltraData-Math-L3/Multi-Style-Synthetic/ | 3,750,000,000 | 375,000 |
| math_qa | openbmb/UltraData-Math | fe10db8efd35597fd7fcff8ff576b5ec4ea5ff87 | data/UltraData-Math-L3/QA-Synthetic/ | 3,750,000,000 | 375,000 |
| math_textbook | openbmb/UltraData-Math | fe10db8efd35597fd7fcff8ff576b5ec4ea5ff87 | data/UltraData-Math-L3/Textbook-Exercise-Synthetic/ | 3,750,000,000 | 375,000 |
| code_l2 | openbmb/UltraData-Code | 85182d829f2ce7ea07cca72ebfc509deea1d9f5f | data/UltraData-Code-L2/py/ | 2,500,000,000 | 250,000 |
| code_l3 | openbmb/UltraData-Code | 85182d829f2ce7ea07cca72ebfc509deea1d9f5f | data/UltraData-Code-L3/py/ | 2,500,000,000 | 250,000 |
| nemotron_actual | CommonCrawl/Nemotron-CC | index-sha256:d8201c1e05b5e9fecef7678457c172f98a19a25f1dd656cdba1f937e7a29e68e | contrib/Nemotron/Nemotron-CC/data-jsonl/quality=high/kind=actual/kind2=actual/ | 5,254,156,769 | 525,415 |
| nemotron_qa | CommonCrawl/Nemotron-CC | index-sha256:d8201c1e05b5e9fecef7678457c172f98a19a25f1dd656cdba1f937e7a29e68e | contrib/Nemotron/Nemotron-CC/data-jsonl/quality=high/kind=synthetic/kind2=diverse_qa_pairs/ | 4,745,843,231 | 474,585 |
The selectors and fetched-file receipts are recorded in corpus-v2/plan.json and corpus-v2/source-manifest.json. Four UltraData-Math L3 synthetic styles divide the 15B math allocation equally. The direct Nemotron component uses the original Common Crawl-hosted Nemotron-CC high-quality actual and diverse-QA variants; it is not Nemotron-CC-Math. UltraData-Math acknowledges NVIDIA, MegaMath, and FineMath among its seed resources, so removing a direct FineMath stream does not remove all FineMath ancestry. See corpus-v2/ATTRIBUTION.md and corpus-v2/provenance/source-research.md.
## Format
- corpus-v2/tokens.bin: little-endian unsigned 16-bit token IDs (<u2). A stored document begins with ID 2,047.
- corpus-v2/<split>-<source>.idx: little-endian unsigned 64-bit triples (offset, selected_length, full_length), 24 bytes per selected document. Split is train or validation. Offsets and lengths count token positions.
- Matching .digests files contain 32-byte SHA-256 content identities. Matching .origins files contain (source_file_index: <u4, source_row_index: <u8) records (12 bytes per document).
- The manifest binds each index, digest and origin sidecar. Use only the source files named by the manifest; do not treat every physical token position as training data.
## Verification checks and results
Publication was independently checked on **30 September 2026 (UTC)**. The local finalizer and public-repository verification passed the checks below.
| Check | Recorded result |
| --- | --- |
| Selected positions and source quotas | PASS — exactly 100,000,000,000 training and 10,000,000 validation positions; all 11 component quotas matched. |
| Full local output hashes | PASS — SHA-256 recomputed for all 69 builder output files. |
| Token range and document boundaries | PASS — stored token IDs checked against the 2,048-ID vocabulary; boundary ID 2,047 checked for all 96,948,408 selected documents. |
| Physical token span coverage | PASS — full document spans cover the token file without overlaps, gaps or an unreferenced tail. |
| Digest, split and source provenance | PASS — all recorded digest-to-split assignments checked; source download receipts bound to the pinned plan. |
| Fast tokenizer parity | PASS — 9,479 sampled Splintr/reference comparisons matched. This was sampled parity, not dual-tokenization of every record. |
| Production loader | PASS — 311,296-position batches and exact state save/resume. |
| Held-out loader coverage | PASS — one complete 10,000,000-position validation traversal. |
| Published corpus integrity | PASS — all 131 corpus files, totaling 207,186,190,336 bytes, matched remote size and content-hash evidence at immutable revisions. |
| Completion receipt and legacy retirement | PASS — public receipt matched the local receipt; all 180 authorized legacy paths were absent at the retirement revision. |
Verified data revision: [`76c88a54094708e776e688d6f745c92b540167e7`](https://huggingface.co/datasets/Jonnester/purrence-hyperbullet-100b/tree/76c88a54094708e776e688d6f745c92b540167e7/corpus-v2). Retirement revision: [`41d30eaf4fe4d2b26650dfcee5fced52d6a2fb41`](https://huggingface.co/datasets/Jonnester/purrence-hyperbullet-100b/tree/41d30eaf4fe4d2b26650dfcee5fced52d6a2fb41). Large LFS files were checked against remote SHA-256 and size metadata. For small Git files, their local SHA-256 was checked against the plan and their Git blob SHA-1 against the remote object. The independent publication check fetched the public receipt and metadata anonymously; it did not redownload and rehash the entire 207 GB payload.
Evidence: [finalizer results](https://huggingface.co/datasets/Jonnester/purrence-hyperbullet-100b/blob/76c88a54094708e776e688d6f745c92b540167e7/corpus-v2/verification.json), [loader readiness](https://huggingface.co/datasets/Jonnester/purrence-hyperbullet-100b/blob/76c88a54094708e776e688d6f745c92b540167e7/corpus-v2/readiness.json), [completion receipt](https://huggingface.co/datasets/Jonnester/purrence-hyperbullet-100b/resolve/41d30eaf4fe4d2b26650dfcee5fced52d6a2fb41/UPLOAD_COMPLETE.json), and [independent publication-check results](https://huggingface.co/datasets/Jonnester/purrence-hyperbullet-100b/blob/main/reports/corpus-v2-validation-20260930/publication-verification.json). The receipt covers **132 files / 207,186,195,962 bytes** when the original dataset card is included. It pins that original card at the data revision; this later documentation update does not change the corpus payload or the receipt. Legacy retirement removes files from the latest tree, not from historical commits.
### Data screening results
Examined **98,192,743 source records** and retained **96,948,408 documents** (96,938,586 training and 9,822 validation). Hygiene, exact deduplication and split/evaluation safeguards excluded **1,224,235 records (1.2468%)**. Another **13,673** records were skipped because their assigned split quota was full, and **6,427** buffered records were not selected after quotas filled; these are not quality failures.
| Recorded exclusion reason | Records |
| --- | ---: |
| Exact content duplicates | 416,177 |
| Environment, VCS, standard-library or binary copies | 252,315 |
| English-language checks | 275,360 |
| Too short | 125,602 |
| Python 3.11 syntax check failures | 81,431 |
| Excessive repetition | 39,566 |
| Excess markup, URLs or UI/commercial boilerplate | 15,804 |
| Tokenizer fit/heldout conflicts | 5,971 |
| Oversized records | 5,250 |
| Broken encoding or control characters | 3,968 |
| Conspicuously unfinished code answers | 1,809 |
| Embedded or chunked encoded payloads | 607 |
| Evaluation prompt matches | 375 |
Counts are source records, not tokens, and each excluded record is counted under its first failing check. Exact rejected-token volume was not measured. The 375 evaluation-prompt matches comprised **258 ARC-Easy, 94 ARC-Challenge and 23 PIQA** records; zero recorded matches for the other tasks does not establish absence of contamination. Python syntax screening is not execution or correctness validation.
[Download the complete cleaning counts and per-source breakdown](https://huggingface.co/datasets/Jonnester/purrence-hyperbullet-100b/blob/main/reports/corpus-v2-validation-20260930/cleaning-results.json). Exclusion rates apply only to consumed records and use source-specific rules; they are not a quality ranking of upstream datasets. Mathematical/factual correctness, proof validity, code execution, semantic deduplication and exhaustive benchmark decontamination were not established by these checks.
## Quality and limitations
This corpus was structurally screened, exact-content deduplicated, assigned to train/validation using content and available source identity, and checked against tokenizer-fit and exact evaluation-prompt exclusions. The builder and finalizer receipts describe these checks. They do not certify semantic correctness, mathematical proof validity, QA entailment, semantic near-duplicate removal, complete identity isolation, or complete benchmark decontamination. The accepted UltraData-Math L3 sample audit found mathematical and answer errors; see corpus-v2/provenance/AUDIT.md. One-shard samples are bounded evidence and are not a corpus-wide error-rate estimate.
No raw source text or sample previews are included as separate files. Upstream terms remain applicable to source content; this combined tokenized derivative does not impose a new blanket license. Consult each upstream card/license and the detailed source-terms discussion in corpus-v2/provenance/source-research.md.
This repository contains no model weights, training code package, or model card. This is a dataset card.