Add files using upload-large-folder tool

README.md

@@ -0,0 +1,140 @@
+---
+license:
+- cc-by-4.0
+- odc-by
+language:
+- en
+pretty_name: ShellElastic pretraining data
+size_categories:
+- 10M<n<100M
+task_categories:
+- text-generation
+tags:
+- shellelastic
+- elastic-pretraining
+- pretraining
+- tokenized
+- webdataset
+viewer: false
+---
+
+# ShellElastic pretraining data (Table 2)
+
+The tokenized training and validation data of the Table 2 runs of *ShellElastic: Scalable Elastic Pretraining with Simultaneous Maximal Updates and Hyperparameter Transfer* (Jin et al.; Rutgers University and Adobe Research). As described in Section 4.1 of the paper,
+training uses [DCLM-Baseline](https://huggingface.co/datasets/mlfoundations/dclm-baseline-1.0) (Li et al., 2024) and validation held-out [C4](https://huggingface.co/datasets/allenai/c4) text (Raffel et al.,
+2020), tokenized with the [GPT-NeoX-20B tokenizer](https://huggingface.co/EleutherAI/gpt-neox-20b) (Black et al., 2022) and packed into sequences of 2,049
+tokens. This repository holds exactly the shards that the runs read, in the layout that the code of the paper reads.
+The final checkpoints of these runs are in the model repositories [Can111/ShellElastic-119M-285M](https://huggingface.co/Can111/ShellElastic-119M-285M), [Can111/ShellElastic-375M-906M](https://huggingface.co/Can111/ShellElastic-375M-906M) and [Can111/ShellElastic-675M-1.69B](https://huggingface.co/Can111/ShellElastic-675M-1.69B) and the baseline repositories
+`Can111/ShellElastic-baseline-*`.
+
+## Contents
+
+| Path | Contents | Size | Sequences | Used for |
+|---|---|---|---|---|
+| `dclm-baseline/` | 3,136 shards (`*.tar`) and `manifest.jsonl` | 141.32 GB | 25,690,112 | training |
+| `c4-val/` | 11 shards (`shard-*.tar`) | 1.12 GB | 84,999 | validation |
+| `shards.tsv` | the training shards in the order the runs read them, with the sequences read from each at every scale | | | |
+
+## Format
+
+Every shard is a tar file in WebDataset format, with one sequence of 2,049 token ids per sample:
+
+- `dclm-baseline/`: 8,192 samples per shard, each a gzip-compressed JSON list of token ids (`<key>.json.gz`);
+- `c4-val/`: each sample a Python list literal (`<key>.txt`) of token ids, in which the end of a document is written
+  as the string `'<|endoftext|>'`; the code reads it as token id 0, the `<|endoftext|>` token of the tokenizer.
+
+Token ids index the 50,432-entry vocabulary of the GPT-NeoX-20B tokenizer (Table 7 of the paper). A training example
+takes the first 2,048 tokens of a sample as its inputs and the last 2,048 as its next-token targets.
+
+`dclm-baseline/manifest.jsonl` has one line `{"shard": ..., "num_sequences": ...}` for each of the 22,981 shards
+of the tokenized DCLM-Baseline subset that these shards come from. Only the 3,136 shards that the paper's runs read
+are included; the manifest is kept whole because the order in which the code reads the shards is a seeded shuffle of
+all of its entries.
+
+## Usage
+
+The configs of the code repository read the training data from `data/dclm-baseline` and the validation data from
+`data/c4-val`. From the root of the code repository:
+
+```python
+from huggingface_hub import snapshot_download
+
+snapshot_download("Can111/ShellElastic-pretraining-data", repo_type="dataset", local_dir="data")
+```
+
+To download only the shards that the runs of one scale read (here 119M–285M; the columns `read_375M-906M` and
+`read_675M-1.69B` of `shards.tsv` select the other two scales):
+
+```python
+import csv
+from huggingface_hub import hf_hub_download, snapshot_download
+
+repo = "Can111/ShellElastic-pretraining-data"
+with open(hf_hub_download(repo, "shards.tsv", repo_type="dataset")) as f:
+    shards = [r["shard"] for r in csv.DictReader(f, delimiter="\t") if int(r["read_119M-285M"]) > 0]
+snapshot_download(repo, repo_type="dataset", local_dir="data",
+                  allow_patterns=[f"dclm-baseline/{s}.tar" for s in shards]
+                  + ["dclm-baseline/manifest.jsonl", "c4-val/*"])
+```
+
+## Which shards the runs read
+
+With the settings of the paper (8 GPUs with 8 data-loader workers each, data seed 1), the code shuffles the entries
+of `manifest.jsonl` with the seed and splits the shards of that order among the GPUs and workers. All runs at a scale
+read the same data in the same order, and the three scales read the first shards of the same order:
+
+| Scale | Steps × batch | Shards read | Sequences read | Training tokens | Size of the shards |
+|---|---|---|---|---|---|
+| 119M–285M | 25,000 × 128 | the first 448 | 3,200,000 | 6.55B | 20.19 GB |
+| 375M–906M | 50,000 × 256 | the first 1,600 | 12,800,000 | 26.2B | 72.10 GB |
+| 675M–1.69B | 100,000 × 256 | all 3,136 | 25,600,000 | 52.4B | 141.32 GB |
+
+The runs read the last 64 shards of each scale only in part; `shards.tsv` gives the number of sequences read from
+every shard. With other numbers of GPUs or workers the shards are split differently, and the longest runs can then
+need shards that are not included here.
+
+## Limitations
+
+The data are filtered web text. They can contain incorrect, biased or offensive content, and are meant for research
+on language model pretraining.
+
+## License
+
+- `dclm-baseline/`: a subset of [DCLM-Baseline](https://huggingface.co/datasets/mlfoundations/dclm-baseline-1.0) (Li et al., 2024), which is distributed under
+  [CC-BY-4.0](https://creativecommons.org/licenses/by/4.0/) and derived from Common Crawl. Changes: the text was tokenized with the GPT-NeoX-20B
+  tokenizer, packed into sequences of 2,049 tokens and shuffled into shards of 8,192 sequences, and only the
+  shards that the paper's runs read are included. These data are distributed under CC-BY-4.0.
+- `c4-val/`: held-out text of [C4](https://huggingface.co/datasets/allenai/c4) (Raffel et al., 2020), which is distributed under [ODC-BY](https://opendatacommons.org/licenses/by/1-0/)
+  together with the [Common Crawl terms of use](https://commoncrawl.org/terms-of-use/), tokenized and packed in the same way. These data are
+  distributed under ODC-BY, and the Common Crawl terms of use apply to them as well.
+- The token ids refer to the [GPT-NeoX-20B tokenizer](https://huggingface.co/EleutherAI/gpt-neox-20b) (Apache-2.0), which is not bundled here.
+
+The underlying web text belongs to its authors. DCLM-Baseline is intended for research use (see its dataset card).
+
+## Citation
+
+```bibtex
+@article{jin2026shellelastic,
+  title   = {ShellElastic: Scalable Elastic Pretraining with Simultaneous Maximal Updates and Hyperparameter Transfer},
+  author  = {Jin, Can and Kang, Yan and Gong, Yifan and Ge, Chongjian and Xu, Mingze and Xiong, Yuanjun and Metaxas, Dimitris N. and Peng, Hongwu},
+  journal = {arXiv preprint arXiv:XXXX.XXXXX},
+  year    = {2026}
+}
+
+@article{li2024datacomplm,
+  title   = {DataComp-LM: In search of the next generation of training sets for language models},
+  author  = {Jeffrey Li and Alex Fang and Georgios Smyrnis and Maor Ivgi and Matt Jordan and Samir Gadre and Hritik Bansal and Etash Guha and Sedrick Keh and Kushal Arora and Saurabh Garg and Rui Xin and Niklas Muennighoff and Reinhard Heckel and Jean Mercat and Mayee Chen and Suchin Gururangan and Mitchell Wortsman and Alon Albalak and Yonatan Bitton and Marianna Nezhurina and Amro Abbas and Cheng-Yu Hsieh and Dhruba Ghosh and Josh Gardner and Maciej Kilian and Hanlin Zhang and Rulin Shao and Sarah Pratt and Sunny Sanyal and Gabriel Ilharco and Giannis Daras and Kalyani Marathe and Aaron Gokaslan and Jieyu Zhang and Khyathi Chandu and Thao Nguyen and Igor Vasiljevic and Sham Kakade and Shuran Song and Sujay Sanghavi and Fartash Faghri and Sewoong Oh and Luke Zettlemoyer and Kyle Lo and Alaaeldin El-Nouby and Hadi Pouransari and Alexander Toshev and Stephanie Wang and Dirk Groeneveld and Luca Soldaini and Pang Wei Koh and Jenia Jitsev and Thomas Kollar and Alexandros G. Dimakis and Yair Carmon and Achal Dave and Ludwig Schmidt and Vaishaal Shankar},
+  journal = {arXiv preprint arXiv:2406.11794},
+  year    = {2024}
+}
+
+@article{raffel2020exploring,
+  title   = {Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer},
+  author  = {Colin Raffel and Noam Shazeer and Adam Roberts and Katherine Lee and Sharan Narang and Michael Matena and Yanqi Zhou and Wei Li and Peter J. Liu},
+  journal = {Journal of Machine Learning Research},
+  volume  = {21},
+  number  = {140},
+  pages   = {1--67},
+  year    = {2020}
+}
+```

c4-val/shard-0000010.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:834bc6d9f8f1413d9b67bc9e2188ca72da02fa02a4c7a2fd9980ea21517377e4
+size 40586240

dclm-baseline/0_00000095.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5e6c49abf7eca7827bf1722434186f19e1af918c521ca739c9608b9ab6f2b0cb
+size 45045760

dclm-baseline/0_00000468.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c8a30bd2d9c341d28303912e14b70780d8377ad06f779904265099e04369e35d
+size 45035520

dclm-baseline/0_00000530.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:39c7b8013e2022828db5c33928a6c14d690fe7f313caa19703faabdd71c4f7e1
+size 45045760

dclm-baseline/0_00000622.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:682e556795285d6126d545752243699558c6fda39638101725238a96a674fbe2
+size 45066240

dclm-baseline/0_00001196.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:df11a82f2987a53eb7f0be8cd91063b6f2414d3a6fcf05437523762ad5179417
+size 45045760

dclm-baseline/0_00001735.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6333c1f9572875c529cda95240dedd9fcb277d847cb04a327a2734be40635dca
+size 45076480

dclm-baseline/0_00001876.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:993136dde0b2077e1cdf4edcd838fa3f3dec77ecc784a85eaa919b062a9c27a9
+size 45066240

dclm-baseline/0_00001910.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a4214fd40476a76079ec07df526ffc88b536153473fbb71a48bc8d796ea09981
+size 45066240

dclm-baseline/0_00002125.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9baa871eb50d997c6fee9bf46f5e6ca8278aa3eb9558645f21a083e0d539f129
+size 45035520

dclm-baseline/1_00000204.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f37ef4ce9ba842fba0ab4556daf9f123faf97d12ee8e7b897f372ccb5713fc50
+size 45035520

dclm-baseline/1_00001309.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:597bcb5523871e70737d7303e52e52a102a80331b4d67ae7fa2bceef05d6a9ce
+size 45066240

dclm-baseline/1_00001434.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01add11e89a7aa17e881a882eecb31b32a656b899169ff252c0f8cb6c51301b1
+size 45066240

dclm-baseline/2_00000023.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:409c05ebc60495e0e126628200600f5fbf8c7876444a092774904b0159d9536c
+size 45066240

dclm-baseline/2_00000116.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f4bd509342309c6161700a3a057252897478a67cae561927534b12b990e4ca66
+size 45035520

dclm-baseline/2_00000245.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae94b6615c04f3f663f25d7a1b29b56c9d5fac078e12b5cd1922f27a4232c95a
+size 45076480

dclm-baseline/2_00000475.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:50b12a98dec1332a00a9e160e8a3c47398bfd6650a2011dbb12d871197c4ddfa
+size 45045760

dclm-baseline/2_00000510.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:776193265efca2373dc445ee30f8b2b315d3a244c8c9488668abe748149e8f0f
+size 45056000

dclm-baseline/2_00001134.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b0a80321e81e1cef2078f940b03ded1d82aa3d8df8de5a1a05ce59dd18a77791
+size 45045760

dclm-baseline/2_00001824.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5b52fac475ad7a61a79ef7723e1b7581769da995798993331a0e134b534904bc
+size 45076480

dclm-baseline/2_00002047.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:668d0dff76f88b6078e2a9e146fa782a84fb3b1b5f87ab8309eaea92f3cd0a6a
+size 45045760

dclm-baseline/3_00000225.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b273fad7620ffe98ca1beb71dc960492b10ac6af3744771f78507bc5bcc7a7fb
+size 45045760

dclm-baseline/3_00000445.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:54dde693cd0ef486312411b3529bea02c87909cb5340566c6afcd5af3cf2a0ab
+size 45076480

dclm-baseline/3_00001237.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd983d27b6ab7a3161d4d1dbb5bd47d883a0d3e3f88ddc89f9495a8efb0b7167
+size 45096960

dclm-baseline/3_00001472.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:933bd551a264d3621933142005ec231a2c1dbea8b7384312c3914ea2c8a2120b
+size 45045760

dclm-baseline/3_00001590.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:27100c31af9b0fe284b70693b7b1a6dffa487b03d24ce2ec7aae5cffdb29415e
+size 45066240

dclm-baseline/4_00000406.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7e50e38b519371b8bb32a79b41633a14d80d70cc90c5c6272d0a582d311e053e
+size 45066240

dclm-baseline/4_00000671.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82bb046a240cad0834ac2b2b5386f37580cd9f0f84f58d3e96c18cc4a7ad2cda
+size 45086720

dclm-baseline/4_00001757.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:904bf9875df1fb4aad65e316afd633619750acd497edc368fc0a700a2d527dad
+size 45056000

dclm-baseline/4_00001928.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7c9e8519c8c55aaccdb2737ebfc95c6de8d877691e4662be52b4194a3d1128db
+size 45056000

dclm-baseline/4_00001968.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c8e9e2f12e2ee0a78b6c708813dd0402cf1399bb0b3a75f53d5ff8a557049fe
+size 45066240

dclm-baseline/5_00000502.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3071e1733e716bcd7c6bb593670faf16ca9dc877f55042ea80a5fce63e370fc9
+size 45045760

dclm-baseline/5_00001070.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ef1e089d1beaf3653c2d446b46fdd2e38593d2aefded4f2296b713e330293419
+size 45066240

dclm-baseline/5_00001870.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:39c16523e2d7de1b788ff9bababb3e7821db682046e3d80b15b5fed6568a5ec4
+size 45056000

dclm-baseline/5_00002160.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c3ac991ef0b1012e309e052e90459353189db8e485e3384a295dd1f49cdd016
+size 45025280

dclm-baseline/6_00000408.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c73ff5afe11a6aea930c02d1223756d415e5765560f6805ce0a249bb50d04a22
+size 45056000

dclm-baseline/6_00000905.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6a32f31c7b9fb44561d7cbd4be35768916d3988cf3511177ef63ad592a06c9e3
+size 45056000

dclm-baseline/6_00001493.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b0e2791aced55d90f26849d4b94e147360c6c09d5e1c18ecf1174e5b7b0d4e68
+size 45035520

dclm-baseline/7_00000328.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca19f9769c47821541fd3abe35bd08a3ce84377ef9e01d571afdae66ca461153
+size 45076480

dclm-baseline/7_00000522.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cf4b153ab78f794a7ad5ba6b75aae52b2ea1b6fa6b49e9fe9f55410d558f29ee
+size 45076480

dclm-baseline/7_00000689.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:21e17995e9fc4ffcdd29827a8aada40b13127dcee1c50700fcae94501394541f
+size 45056000

dclm-baseline/7_00001280.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b4b30fe724eb8c441ad8b8559269b67c47bee5ca3eaf3cd51c58da023155640
+size 45066240

dclm-baseline/8_00000336.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f283c4109d8ab4a183ffb18dd0449808fc21add91da2bb5d9a2ec740431b2ff
+size 45107200

dclm-baseline/8_00001600.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5e6654e99308496412a844a342f0f0c0831c8685b46974729a3c413fabe4ba03
+size 45035520

dclm-baseline/9_00000549.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a5351bdbdb2e7dc6b022659cc2718139b93d8b9eb7325d9eed9f39455c80140c
+size 45086720

dclm-baseline/9_00001517.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:769b9ee5c9ce32038263ca2fbeb2631a51787f02d02804c21923e061d58653a1
+size 45056000

dclm-baseline/9_00001825.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3a62dc2493c312e2b014a8bf7e49a475c75344f97aa321fa45bb367dd488a252
+size 45086720