Dataset Viewer
The dataset viewer is not available because its heuristics could not detect any supported data files. You can try uploading some data files, or configuring the data files location manually.
YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
olmocr3-books-remix-1T-stack
Megatron-LM IndexedDataset shards of a 1T-token document-level remix of the
olmo3 "books" corpus (re-sampled with target token budgets per source, shuffled
document order), packed with step 04A stack (documents concatenated, EOS
between documents, no BOS).
- Format: Megatron-LM IndexedDataset — paired
data_<N>_000000.bin(token ids,uint32)data_<N>_000000.idx(index). 96 shards, ~40 GiB each.
- Total: 1,017,001,409,160 tokens (~1017B), 3.71 TiB on disk.
- Tokenizer: DeepSeek-V3 tokenizer (
tokenizer.json). EOS appended to every document. - Generation:
DataPackingpipeline of the DataEng repo — step 03B token mixture (quick-mix) → step 04A stack (--bin-size-gb 50,--no-bos-token). - Verification: every file was verified byte-for-byte (size) against the
source cluster after upload. To verify locally, re-count tokens with the
Megatron
fast_count_indexed_tokenscheck and compare against the header total above.
- Downloads last month
- 155