glennmatlin's picture
Add sampling metadata and data card
c22bb58 verified
---
license: odc-by
task_categories:
- text-generation
language:
- en
---
# dolma3_6T_sample_500_docs
Materialized working sample from the Dolma3 6T deduplicated corpus.
## Parameters
| Parameter | Value |
|-----------|-------|
| Sampling mode | 500 docs/bin |
| Seed | 42 |
## Counts
| Metric | Value |
|--------|-------|
| Total documents | 287,936 |
| Total tokens | 538,830,013 |
| Bins covered | 576/576 |
| Bins underfilled | 1 |
## Format
JSONL files compressed with zstandard under `data/`.
Each record contains the original Dolma document fields (id, text, metadata).
Sampling metadata files:
- `working_sample_manifest.parquet` - sampled doc_ids with bin assignments
- `sample_contract.json` - aggregate sampling contract
- `bin_summary.csv` - per-bin fill rates (576 bins = 24 topics x 24 formats)