add citations!

bibliography.bib

@@ -0,0 +1,524 @@
+@misc{refinedweb,
+      title={The RefinedWeb Dataset for Falcon LLM: Outperforming Curated Corpora with Web Data, and Web Data Only},
+      author={Guilherme Penedo and Quentin Malartic and Daniel Hesslow and Ruxandra Cojocaru and Alessandro Cappelli and Hamza Alobeidli and Baptiste Pannier and Ebtesam Almazrouei and Julien Launay},
+      year={2023},
+      eprint={2306.01116},
+      archivePrefix={arXiv},
+      primaryClass={cs.CL}
+}
+@article{redpajama-v1,
+  author = {Together Computer},
+  title = {RedPajama-Data: An Open Source Recipe to Reproduce LLaMA training dataset},
+  month = {April},
+  year = 2023,
+  url = {https://github.com/togethercomputer/RedPajama-Data/tree/rp_v1}
+}
+@article{redpajama-v2,
+  author = {Together Computer},
+  title = {RedPajama-Data-v2: an Open Dataset with 30 Trillion Tokens for Training Large Language Models},
+  month = {October},
+  year = 2023,
+  url = {https://github.com/togethercomputer/RedPajama-Data}
+}
+@article{dolma,
+  title = {Dolma: an Open Corpus of Three Trillion Tokens for Language Model Pretraining Research},
+  author={
+    Luca Soldaini and Rodney Kinney and Akshita Bhagia and Dustin Schwenk and David Atkinson and
+    Russell Authur and Ben Bogin and Khyathi Chandu and Jennifer Dumas and Yanai Elazar and
+    Valentin Hofmann and Ananya Harsh Jha and Sachin Kumar and Li Lucy and Xinxi Lyu and
+    Nathan Lambert and Ian Magnusson and Jacob Morrison and Niklas Muennighoff and Aakanksha Naik and
+    Crystal Nam and Matthew E. Peters and Abhilasha Ravichander and Kyle Richardson and Zejiang Shen and
+    Emma Strubell and Nishant Subramani and Oyvind Tafjord and Pete Walsh and Luke Zettlemoyer and
+    Noah A. Smith and Hannaneh Hajishirzi and Iz Beltagy and Dirk Groeneveld and Jesse Dodge and Kyle Lo
+  },
+  year = {2024},
+  journal={arXiv preprint},
+}
+@article{fineweb,
+  author = {Guilherme Penedo and Hynek Kydlíček and Loubna Ben Allal and Anton Lozhkov and Colin Raffel and Leandro Werra and Thomas Wolf},
+  title = {🍷 FineWeb: decanting the web for the finest text data at scale},
+  month = {May},
+  year = 2024,
+  url = {https://huggingface.co/spaces/HuggingFaceFW/blogpost-fineweb-v1},
+}
+@misc{cerebras2023slimpajama,
+  author = {Soboleva, Daria and Al-Khateeb, Faisal and Myers, Robert and Steeves, Jacob R and Hestness, Joel and Dey, Nolan},
+  title = {SlimPajama: A 627B token cleaned and deduplicated version of RedPajama},
+  month = {June},
+  year = 2023,
+  url = {https://huggingface.co/datasets/cerebras/SlimPajama-627B},
+}
+@article{thepile,
+  title={The {P}ile: An 800{GB} dataset of diverse text for language modeling},
+  author={Gao, Leo and Biderman, Stella and Black, Sid and Golding, Laurence and Hoppe, Travis and Foster, Charles and Phang, Jason and He, Horace and Thite, Anish and Nabeshima, Noa and others},
+  journal={arXiv preprint arXiv:2101.00027},
+  year={2020}
+}
+@misc{c4,
+      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},
+      year={2023},
+      eprint={1910.10683},
+      archivePrefix={arXiv},
+      primaryClass={cs.LG}
+}
+@article{dclm,
+  author       = {Jeffrey Li and
+                  Alex Fang and
+                  Georgios Smyrnis and
+                  Maor Ivgi and
+                  Matt Jordan and
+                  Samir Yitzhak Gadre and
+                  Hritik Bansal and
+                  Etash Kumar 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 M. Pratt and
+                  Sunny Sanyal and
+                  Gabriel Ilharco and
+                  Giannis Daras and
+                  Kalyani Marathe and
+                  Aaron Gokaslan and
+                  Jieyu Zhang and
+                  Khyathi Raghavi Chandu and
+                  Thao Nguyen and
+                  Igor Vasiljevic and
+                  Sham M. 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},
+  title        = {DataComp-LM: In search of the next generation of training sets for
+                  language models},
+  journal      = {CoRR},
+  volume       = {abs/2406.11794},
+  year         = {2024},
+  url          = {https://doi.org/10.48550/arXiv.2406.11794},
+  doi          = {10.48550/ARXIV.2406.11794},
+  eprinttype    = {arXiv},
+  eprint       = {2406.11794},
+  timestamp    = {Mon, 02 Sep 2024 16:44:37 +0200},
+  biburl       = {https://dblp.org/rec/journals/corr/abs-2406-11794.bib},
+  bibsource    = {dblp computer science bibliography, https://dblp.org}
+}
+@article{gopher,
+  author       = {Jack W. Rae and
+                  Sebastian Borgeaud and
+                  Trevor Cai and
+                  Katie Millican and
+                  Jordan Hoffmann and
+                  H. Francis Song and
+                  John Aslanides and
+                  Sarah Henderson and
+                  Roman Ring and
+                  Susannah Young and
+                  Eliza Rutherford and
+                  Tom Hennigan and
+                  Jacob Menick and
+                  Albin Cassirer and
+                  Richard Powell and
+                  George van den Driessche and
+                  Lisa Anne Hendricks and
+                  Maribeth Rauh and
+                  Po{-}Sen Huang and
+                  Amelia Glaese and
+                  Johannes Welbl and
+                  Sumanth Dathathri and
+                  Saffron Huang and
+                  Jonathan Uesato and
+                  John Mellor and
+                  Irina Higgins and
+                  Antonia Creswell and
+                  Nat McAleese and
+                  Amy Wu and
+                  Erich Elsen and
+                  Siddhant M. Jayakumar and
+                  Elena Buchatskaya and
+                  David Budden and
+                  Esme Sutherland and
+                  Karen Simonyan and
+                  Michela Paganini and
+                  Laurent Sifre and
+                  Lena Martens and
+                  Xiang Lorraine Li and
+                  Adhiguna Kuncoro and
+                  Aida Nematzadeh and
+                  Elena Gribovskaya and
+                  Domenic Donato and
+                  Angeliki Lazaridou and
+                  Arthur Mensch and
+                  Jean{-}Baptiste Lespiau and
+                  Maria Tsimpoukelli and
+                  Nikolai Grigorev and
+                  Doug Fritz and
+                  Thibault Sottiaux and
+                  Mantas Pajarskas and
+                  Toby Pohlen and
+                  Zhitao Gong and
+                  Daniel Toyama and
+                  Cyprien de Masson d'Autume and
+                  Yujia Li and
+                  Tayfun Terzi and
+                  Vladimir Mikulik and
+                  Igor Babuschkin and
+                  Aidan Clark and
+                  Diego de Las Casas and
+                  Aurelia Guy and
+                  Chris Jones and
+                  James Bradbury and
+                  Matthew J. Johnson and
+                  Blake A. Hechtman and
+                  Laura Weidinger and
+                  Iason Gabriel and
+                  William Isaac and
+                  Edward Lockhart and
+                  Simon Osindero and
+                  Laura Rimell and
+                  Chris Dyer and
+                  Oriol Vinyals and
+                  Kareem Ayoub and
+                  Jeff Stanway and
+                  Lorrayne Bennett and
+                  Demis Hassabis and
+                  Koray Kavukcuoglu and
+                  Geoffrey Irving},
+  title        = {Scaling Language Models: Methods, Analysis {\&} Insights from
+                  Training Gopher},
+  journal      = {CoRR},
+  volume       = {abs/2112.11446},
+  year         = {2021},
+  url          = {https://arxiv.org/abs/2112.11446},
+  eprinttype    = {arXiv},
+  eprint       = {2112.11446},
+  timestamp    = {Sat, 02 Dec 2023 13:23:51 +0100},
+  biburl       = {https://dblp.org/rec/journals/corr/abs-2112-11446.bib},
+  bibsource    = {dblp computer science bibliography, https://dblp.org}
+}
+@article{radford2019language,
+  title={Language Models are Unsupervised Multitask Learners},
+  author={Radford, Alec and Wu, Jeff and Child, Rewon and Luan, David and Amodei, Dario and Sutskever, Ilya},
+  year={2019}
+}
+@inproceedings{barbaresi-2021-trafilatura,
+  title = {Trafilatura: A Web Scraping Library and Command-Line Tool for Text Discovery and Extraction},
+  author = "Barbaresi, Adrien",
+  booktitle = "Proceedings of the Joint Conference of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing: System Demonstrations",
+  pages = "122--131",
+  publisher = "Association for Computational Linguistics",
+  url = "https://aclanthology.org/2021.acl-demo.15",
+  year = 2021,
+}
+@article{joulin2016fasttext,
+  title={FastText.zip: Compressing text classification models},
+  author={Joulin, Armand and Grave, Edouard and Bojanowski, Piotr and Douze, Matthijs and J{\'e}gou, H{\'e}rve and Mikolov, Tomas},
+  journal={arXiv preprint arXiv:1612.03651},
+  year={2016}
+}
+@article{joulin2016bag,
+  title={Bag of Tricks for Efficient Text Classification},
+  author={Joulin, Armand and Grave, Edouard and Bojanowski, Piotr and Mikolov, Tomas},
+  journal={arXiv preprint arXiv:1607.01759},
+  year={2016}
+}
+@misc{penedo2024datatrove,
+  author = {Penedo, Guilherme and Kydlíček, Hynek and Cappelli, Alessandro and Sasko, Mario and Wolf, Thomas},
+  title = {DataTrove: large scale data processing},
+  year = {2024},
+  publisher = {GitHub},
+  journal = {GitHub repository},
+  url = {https://github.com/huggingface/datatrove}
+}
+@misc{chiang2024chatbot,
+      title={Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preference},
+      author={Wei-Lin Chiang and Lianmin Zheng and Ying Sheng and Anastasios Nikolas Angelopoulos and Tianle Li and Dacheng Li and Hao Zhang and Banghua Zhu and Michael Jordan and Joseph E. Gonzalez and Ion Stoica},
+      year={2024},
+      eprint={2403.04132},
+      archivePrefix={arXiv},
+      primaryClass={cs.AI}
+}
+@misc{rae2022scaling,
+      title={Scaling Language Models: Methods, Analysis & Insights from Training Gopher},
+      author={Jack W. Rae and Sebastian Borgeaud and Trevor Cai and Katie Millican and Jordan Hoffmann and Francis Song and John Aslanides and Sarah Henderson and Roman Ring and Susannah Young and Eliza Rutherford and Tom Hennigan and Jacob Menick and Albin Cassirer and Richard Powell and George van den Driessche and Lisa Anne Hendricks and Maribeth Rauh and Po-Sen Huang and Amelia Glaese and Johannes Welbl and Sumanth Dathathri and Saffron Huang and Jonathan Uesato and John Mellor and Irina Higgins and Antonia Creswell and Nat McAleese and Amy Wu and Erich Elsen and Siddhant Jayakumar and Elena Buchatskaya and David Budden and Esme Sutherland and Karen Simonyan and Michela Paganini and Laurent Sifre and Lena Martens and Xiang Lorraine Li and Adhiguna Kuncoro and Aida Nematzadeh and Elena Gribovskaya and Domenic Donato and Angeliki Lazaridou and Arthur Mensch and Jean-Baptiste Lespiau and Maria Tsimpoukelli and Nikolai Grigorev and Doug Fritz and Thibault Sottiaux and Mantas Pajarskas and Toby Pohlen and Zhitao Gong and Daniel Toyama and Cyprien de Masson d'Autume and Yujia Li and Tayfun Terzi and Vladimir Mikulik and Igor Babuschkin and Aidan Clark and Diego de Las Casas and Aurelia Guy and Chris Jones and James Bradbury and Matthew Johnson and Blake Hechtman and Laura Weidinger and Iason Gabriel and William Isaac and Ed Lockhart and Simon Osindero and Laura Rimell and Chris Dyer and Oriol Vinyals and Kareem Ayoub and Jeff Stanway and Lorrayne Bennett and Demis Hassabis and Koray Kavukcuoglu and Geoffrey Irving},
+      year={2022},
+      eprint={2112.11446},
+      archivePrefix={arXiv},
+      primaryClass={cs.CL}
+}
+@misc{lee2022deduplicating,
+      title={Deduplicating Training Data Makes Language Models Better},
+      author={Katherine Lee and Daphne Ippolito and Andrew Nystrom and Chiyuan Zhang and Douglas Eck and Chris Callison-Burch and Nicholas Carlini},
+      year={2022},
+      eprint={2107.06499},
+      archivePrefix={arXiv},
+      primaryClass={cs.CL}
+}
+@misc{carlini2023quantifying,
+      title={Quantifying Memorization Across Neural Language Models},
+      author={Nicholas Carlini and Daphne Ippolito and Matthew Jagielski and Katherine Lee and Florian Tramer and Chiyuan Zhang},
+      year={2023},
+      eprint={2202.07646},
+      archivePrefix={arXiv},
+      primaryClass={cs.LG}
+}
+@misc{touvron2023llama,
+      title={LLaMA: Open and Efficient Foundation Language Models},
+      author={Hugo Touvron and Thibaut Lavril and Gautier Izacard and Xavier Martinet and Marie-Anne Lachaux and Timothée Lacroix and Baptiste Rozière and Naman Goyal and Eric Hambro and Faisal Azhar and Aurelien Rodriguez and Armand Joulin and Edouard Grave and Guillaume Lample},
+      year={2023},
+      eprint={2302.13971},
+      archivePrefix={arXiv},
+      primaryClass={cs.CL}
+}
+@article{jaccard1912distribution,
+  title={The distribution of the flora in the alpine zone. 1},
+  author={Jaccard, Paul},
+  journal={New phytologist},
+  volume={11},
+  number={2},
+  pages={37--50},
+  year={1912},
+  publisher={Wiley Online Library}
+}
+@misc{albalak2024survey,
+      title={A Survey on Data Selection for Language Models},
+      author={Alon Albalak and Yanai Elazar and Sang Michael Xie and Shayne Longpre and Nathan Lambert and Xinyi Wang and Niklas Muennighoff and Bairu Hou and Liangming Pan and Haewon Jeong and Colin Raffel and Shiyu Chang and Tatsunori Hashimoto and William Yang Wang},
+      year={2024},
+      eprint={2402.16827},
+      archivePrefix={arXiv},
+      primaryClass={cs.CL}
+}
+@misc{longpre2023pretrainers,
+      title={A Pretrainer's Guide to Training Data: Measuring the Effects of Data Age, Domain Coverage, Quality, & Toxicity},
+      author={Shayne Longpre and Gregory Yauney and Emily Reif and Katherine Lee and Adam Roberts and Barret Zoph and Denny Zhou and Jason Wei and Kevin Robinson and David Mimno and Daphne Ippolito},
+      year={2023},
+      eprint={2305.13169},
+      archivePrefix={arXiv},
+      primaryClass={cs.CL}
+}
+@misc{wenzek2019ccnet,
+      title={CCNet: Extracting High Quality Monolingual Datasets from Web Crawl Data},
+      author={Guillaume Wenzek and Marie-Anne Lachaux and Alexis Conneau and Vishrav Chaudhary and Francisco Guzmán and Armand Joulin and Edouard Grave},
+      year={2019},
+      eprint={1911.00359},
+      archivePrefix={arXiv},
+      primaryClass={cs.CL}
+}
+@misc{soldaini2024dolma,
+      title={Dolma: an Open Corpus of Three Trillion Tokens for Language Model Pretraining Research},
+      author={Luca Soldaini and Rodney Kinney and Akshita Bhagia and Dustin Schwenk and David Atkinson and Russell Authur and Ben Bogin and Khyathi Chandu and Jennifer Dumas and Yanai Elazar and Valentin Hofmann and Ananya Harsh Jha and Sachin Kumar and Li Lucy and Xinxi Lyu and Nathan Lambert and Ian Magnusson and Jacob Morrison and Niklas Muennighoff and Aakanksha Naik and Crystal Nam and Matthew E. Peters and Abhilasha Ravichander and Kyle Richardson and Zejiang Shen and Emma Strubell and Nishant Subramani and Oyvind Tafjord and Pete Walsh and Luke Zettlemoyer and Noah A. Smith and Hannaneh Hajishirzi and Iz Beltagy and Dirk Groeneveld and Jesse Dodge and Kyle Lo},
+      year={2024},
+      eprint={2402.00159},
+      archivePrefix={arXiv},
+      primaryClass={cs.CL}
+}
+@misc{ouyang2022training,
+      title={Training language models to follow instructions with human feedback},
+      author={Long Ouyang and Jeff Wu and Xu Jiang and Diogo Almeida and Carroll L. Wainwright and Pamela Mishkin and Chong Zhang and Sandhini Agarwal and Katarina Slama and Alex Ray and John Schulman and Jacob Hilton and Fraser Kelton and Luke Miller and Maddie Simens and Amanda Askell and Peter Welinder and Paul Christiano and Jan Leike and Ryan Lowe},
+      year={2022},
+      eprint={2203.02155},
+      archivePrefix={arXiv},
+      primaryClass={cs.CL}
+}
+@misc{hoffmann2022training,
+      title={Training Compute-Optimal Large Language Models},
+      author={Jordan Hoffmann and Sebastian Borgeaud and Arthur Mensch and Elena Buchatskaya and Trevor Cai and Eliza Rutherford and Diego de Las Casas and Lisa Anne Hendricks and Johannes Welbl and Aidan Clark and Tom Hennigan and Eric Noland and Katie Millican and George van den Driessche and Bogdan Damoc and Aurelia Guy and Simon Osindero and Karen Simonyan and Erich Elsen and Jack W. Rae and Oriol Vinyals and Laurent Sifre},
+      year={2022},
+      eprint={2203.15556},
+      archivePrefix={arXiv},
+      primaryClass={cs.CL}
+}
+@misc{muennighoff2023scaling,
+      title={Scaling Data-Constrained Language Models},
+      author={Niklas Muennighoff and Alexander M. Rush and Boaz Barak and Teven Le Scao and Aleksandra Piktus and Nouamane Tazi and Sampo Pyysalo and Thomas Wolf and Colin Raffel},
+      year={2023},
+      eprint={2305.16264},
+      archivePrefix={arXiv},
+      primaryClass={cs.CL}
+}
+@misc{hernandez2022scaling,
+      title={Scaling Laws and Interpretability of Learning from Repeated Data},
+      author={Danny Hernandez and Tom Brown and Tom Conerly and Nova DasSarma and Dawn Drain and Sheer El-Showk and Nelson Elhage and Zac Hatfield-Dodds and Tom Henighan and Tristan Hume and Scott Johnston and Ben Mann and Chris Olah and Catherine Olsson and Dario Amodei and Nicholas Joseph and Jared Kaplan and Sam McCandlish},
+      year={2022},
+      eprint={2205.10487},
+      archivePrefix={arXiv},
+      primaryClass={cs.LG}
+}
+@article{llama3modelcard,
+
+title={Llama 3 Model Card},
+
+author={AI@Meta},
+
+year={2024},
+
+url = {https://github.com/meta-llama/llama3/blob/main/MODEL_CARD.md}
+
+}
+@misc{jiang2024mixtral,
+      title={Mixtral of Experts},
+      author={Albert Q. Jiang and Alexandre Sablayrolles and Antoine Roux and Arthur Mensch and Blanche Savary and Chris Bamford and Devendra Singh Chaplot and Diego de las Casas and Emma Bou Hanna and Florian Bressand and Gianna Lengyel and Guillaume Bour and Guillaume Lample and Lélio Renard Lavaud and Lucile Saulnier and Marie-Anne Lachaux and Pierre Stock and Sandeep Subramanian and Sophia Yang and Szymon Antoniak and Teven Le Scao and Théophile Gervet and Thibaut Lavril and Thomas Wang and Timothée Lacroix and William El Sayed},
+      year={2024},
+      eprint={2401.04088},
+      archivePrefix={arXiv},
+      primaryClass={cs.LG}
+}
+@article{yuan2024self,
+  title={Self-rewarding language models},
+  author={Yuan, Weizhe and Pang, Richard Yuanzhe and Cho, Kyunghyun and Sukhbaatar, Sainbayar and Xu, Jing and Weston, Jason},
+  journal={arXiv preprint arXiv:2401.10020},
+  year={2024}
+}
+@article{verga2024replacing,
+  title={Replacing Judges with Juries: Evaluating LLM Generations with a Panel of Diverse Models},
+  author={Verga, Pat and Hofstatter, Sebastian and Althammer, Sophia and Su, Yixuan and Piktus, Aleksandra and Arkhangorodsky, Arkady and Xu, Minjie and White, Naomi and Lewis, Patrick},
+  journal={arXiv preprint arXiv:2404.18796},
+  year={2024}
+}
+@article{abdin2024phi,
+  title={Phi-3 technical report: A highly capable language model locally on your phone},
+  author={Abdin, Marah and Jacobs, Sam Ade and Awan, Ammar Ahmad and Aneja, Jyoti and Awadallah, Ahmed and Awadalla, Hany and Bach, Nguyen and Bahree, Amit and Bakhtiari, Arash and Behl, Harkirat and others},
+  journal={arXiv preprint arXiv:2404.14219},
+  year={2024}
+}
+@misc{meta2024responsible,
+  title = {Our responsible approach to Meta AI and Meta Llama 3},
+  author = {Meta},
+  year = {2024},
+  url = {https://ai.meta.com/blog/meta-llama-3-meta-ai-responsibility/},
+  note = {Accessed: 2024-05-31}
+}
+@inproceedings{talmor-etal-2019-commonsenseqa,
+    title = "CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge",
+    author = "Talmor, Alon  and
+      Herzig, Jonathan  and
+      Lourie, Nicholas  and
+      Berant, Jonathan",
+    booktitle = "Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)",
+    month = jun,
+    year = "2019",
+    address = "Minneapolis, Minnesota",
+    publisher = "Association for Computational Linguistics",
+    url = "https://aclanthology.org/N19-1421",
+    doi = "10.18653/v1/N19-1421",
+    pages = "4149--4158",
+    archivePrefix = "arXiv",
+    eprint        = "1811.00937",
+    primaryClass  = "cs",
+}
+@inproceedings{zellers-etal-2019-hellaswag,
+    title = "HellaSwag: Can a Machine Really Finish Your Sentence?",
+    author = "Zellers, Rowan  and
+      Holtzman, Ari  and
+      Bisk, Yonatan  and
+      Farhadi, Ali  and
+      Choi, Yejin",
+    editor = "Korhonen, Anna  and
+      Traum, David  and
+      M{\`a}rquez, Llu{\'\i}s",
+    booktitle = "Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics",
+    month = jul,
+    year = "2019",
+    address = "Florence, Italy",
+    publisher = "Association for Computational Linguistics",
+    url = "https://aclanthology.org/P19-1472",
+    doi = "10.18653/v1/P19-1472",
+    pages = "4791--4800",
+    abstract = "Recent work by Zellers et al. (2018) introduced a new task of commonsense natural language inference: given an event description such as {``}A woman sits at a piano,{''} a machine must select the most likely followup: {``}She sets her fingers on the keys.{''} With the introduction of BERT, near human-level performance was reached. Does this mean that machines can perform human level commonsense inference? In this paper, we show that commonsense inference still proves difficult for even state-of-the-art models, by presenting HellaSwag, a new challenge dataset. Though its questions are trivial for humans ({\textgreater}95{\%} accuracy), state-of-the-art models struggle ({\textless}48{\%}). We achieve this via Adversarial Filtering (AF), a data collection paradigm wherein a series of discriminators iteratively select an adversarial set of machine-generated wrong answers. AF proves to be surprisingly robust. The key insight is to scale up the length and complexity of the dataset examples towards a critical {`}Goldilocks{'} zone wherein generated text is ridiculous to humans, yet often misclassified by state-of-the-art models. Our construction of HellaSwag, and its resulting difficulty, sheds light on the inner workings of deep pretrained models. More broadly, it suggests a new path forward for NLP research, in which benchmarks co-evolve with the evolving state-of-the-art in an adversarial way, so as to present ever-harder challenges.",
+}
+@inproceedings{OpenBookQA2018,
+ title={Can a Suit of Armor Conduct Electricity? A New Dataset for Open Book Question Answering},
+ author={Todor Mihaylov and Peter Clark and Tushar Khot and Ashish Sabharwal},
+ booktitle={EMNLP},
+ year={2018}
+}
+@misc{bisk2019piqa,
+      title={PIQA: Reasoning about Physical Commonsense in Natural Language},
+      author={Yonatan Bisk and Rowan Zellers and Ronan Le Bras and Jianfeng Gao and Yejin Choi},
+      year={2019},
+      eprint={1911.11641},
+      archivePrefix={arXiv},
+      primaryClass={cs.CL}
+}
+@misc{sap2019socialiqa,
+      title={SocialIQA: Commonsense Reasoning about Social Interactions},
+      author={Maarten Sap and Hannah Rashkin and Derek Chen and Ronan LeBras and Yejin Choi},
+      year={2019},
+      eprint={1904.09728},
+      archivePrefix={arXiv},
+      primaryClass={cs.CL}
+}
+@misc{sakaguchi2019winogrande,
+      title={WinoGrande: An Adversarial Winograd Schema Challenge at Scale},
+      author={Keisuke Sakaguchi and Ronan Le Bras and Chandra Bhagavatula and Yejin Choi},
+      year={2019},
+      eprint={1907.10641},
+      archivePrefix={arXiv},
+      primaryClass={cs.CL}
+}
+@misc{clark2018think,
+      title={Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge},
+      author={Peter Clark and Isaac Cowhey and Oren Etzioni and Tushar Khot and Ashish Sabharwal and Carissa Schoenick and Oyvind Tafjord},
+      year={2018},
+      eprint={1803.05457},
+      archivePrefix={arXiv},
+      primaryClass={cs.AI}
+}
+@misc{hendrycks2021measuring,
+      title={Measuring Massive Multitask Language Understanding},
+      author={Dan Hendrycks and Collin Burns and Steven Basart and Andy Zou and Mantas Mazeika and Dawn Song and Jacob Steinhardt},
+      year={2021},
+      eprint={2009.03300},
+      archivePrefix={arXiv},
+      primaryClass={cs.CY}
+}
+@misc{mitchell2023measuring,
+      title={Measuring Data},
+      author={Margaret Mitchell and Alexandra Sasha Luccioni and Nathan Lambert and Marissa Gerchick and Angelina McMillan-Major and Ezinwanne Ozoani and Nazneen Rajani and Tristan Thrush and Yacine Jernite and Douwe Kiela},
+      year={2023},
+      eprint={2212.05129},
+      archivePrefix={arXiv},
+      primaryClass={cs.AI}
+}
+@INPROCEEDINGS{6785473,
+  author={Kardes, Hakan and Agrawal, Siddharth and Xin Wang and Ang Sun},
+  booktitle={2014 International Conference on Computing, Networking and Communications (ICNC)},
+  title={CCF: Fast and scalable connected component computation in MapReduce},
+  year={2014},
+  volume={},
+  number={},
+  pages={994-998},
+  keywords={Couplings;Databases;Data mining;Algorithm design and analysis;Social network services;Feature extraction;Cleaning;Transitive Closure;Connected Components;Large Scale Graphs;Hadoop;MapReduce},
+  doi={10.1109/ICCNC.2014.6785473}}

main.py

@@ -1,6 +1,15 @@
 from fasthtml.common import *
 from fasthtml.components import *
-from fasthtml.components import D_title, D_article, D_front_matter, D_contents, D_byline
+from fasthtml.components import (
+    D_title,
+    D_article,
+    D_front_matter,
+    D_contents,
+    D_byline,
+    D_bibliography,
+    D_appendix,
+    D_cite,
+)
 from plotly import graph_objects as go
 from fh_plotly import plotly2fasthtml
 import pandas as pd
@@ -11,6 +20,7 @@ import curated
 import web
 import common
 import results
+from pybtex.database import parse_file
 
 
 app, rt = fast_app(
@@ -28,10 +38,46 @@ app, rt = fast_app(
 )
 
 
+front_matter = """
+<d-front-matter>
+<script id='distill-front-matter' type="text/json">{
+    "title": "",
+    "description": "",
+    "published": "",
+    "affiliation": {},
+    "authors": [
+      {
+        "author":"",
+        "authorURL":""
+      }
+    ],
+    "katex": {
+      "delimiters": [
+        {"left": "$$", "right": "$$", "display": false}
+      ]
+    }
+  }
+</script>
+</d-front-matter>
+"""
+
+
+def read_bibs():
+    bib_data = parse_file("bibliography.bib")
+    cits = []
+    for key in bib_data.entries.keys():
+        cits.append(D_cite(bibtex_key=key))
+    return cits
+
+
+@app.get("/bibliography.bib")
+def get():
+    return FileResponse("bibliography.bib")
+
+
 @app.get("/")
 def main():
     return Div(
-        D_front_matter(),
         D_title(
             H1(
                 "TxT360: the most comprehensive, highest quality, and production ready pretraining dataset",
@@ -44,6 +90,7 @@ def main():
                 cls="main-plot-container l-page",
             ),
         ),
+        Div(D_byline(), NotStr(front_matter), style="display: none;"),
         D_article(
             D_contents(
                 Nav(
@@ -120,148 +167,49 @@ def main():
             ),
             intro(),
         ),
+        D_appendix(D_bibliography(src="bibliography.bib")),
+        Div(*read_bibs(), style="display: none;"),
     )
 
-intro_text = P("Pretraining performant large language models (LLMs) requires trillions of tokens of high quality data. Many prior work, including our previous pretraining projects ",
-            A("Amber-7B", href = "https://huggingface.co/LLM360/Amber"),
-            ", ",
-            A("Crystal-7B", href = "https://huggingface.co/LLM360/CrystalCoder"),
-            ", ",
-            A("K2-65B", href = "https://huggingface.co/LLM360/K2"),
-            " have demonstrated how data curation is a ‘make-or-break’ decision for model quality and capability.",)
 
-intro_list = P("We present TxT360, the Trillion eXtracted Text corpus, a 5.7T token dataset for pretraining projects that:")
-
-intro_list1 = Ol(
-                Li("Curates commonly used pretraining datasets, including all CommonCrawl", style = "margin-bottom: 5px"),
-                Li("Employs carefully selected filters designed for each data source", style = "margin-bottom: 5px"),
-                Li("Provides only unique data elements via globally deduplicated across all datasets", style = "margin-bottom: 5px"),
-                Li("Retains all deduplication metadata for custom upweighting", style = "margin-bottom: 5px"),
-                Li("Is Production ready! Download here [link to HF repo]", style = "margin-bottom: 5px")
+intro_text = P(
+    "Pretraining performant large language models (LLMs) requires trillions of tokens of high quality data. Many prior work, including our previous pretraining projects ",
+    A("Amber-7B", href="https://huggingface.co/LLM360/Amber"),
+    ", ",
+    A("Crystal-7B", href="https://huggingface.co/LLM360/CrystalCoder"),
+    ", ",
+    A("K2-65B", href="https://huggingface.co/LLM360/K2"),
+    " have demonstrated how data curation is a ‘make-or-break’ decision for model quality and capability.",
 )
 
-previous_intro = P("""We are excited to introduce TxT360, a
-                large-scale, comprehensive, and fully transparent
-                dataset designed for Large Language Model (LLM)
-                pre-training. TxT360 is engineered to strike a
-                balance between the quantity and quality of
-                pre-training data, pushing the limit on both
-                fronts. This comprehensive dataset encompasses both
-                expansive web-based data and highly curated data
-                sources, making it one of the most robust LLM
-                pre-training corpora available today.  Our web data
-                component includes 99 snapshots from Common Crawl,
-                amassing 5.7 trillion tokens and occupying 11 TB of
-                disk space in jsonl.gz format. On the curated side,
-                TxT360 integrates one of the most extensive
-                collections of high-quality sources across multiple
-                domains, ensuring diverse and rich content referred
-                to as curated sources, 14 sources across 10
-                domains.  To maintain the highest quality, we
-                meticulously pre-processed the web data to filter
-                out low-quality content and conducted thorough
-                reviews of the curated sources. This process not
-                only unified their formats but also identified and
-                rectified any anomalies. Not only do we 100%
-                open-source our processing scripts, but we also
-                release the details of our data reviews, revealing
-                the decision-making processes behind data selection
-                and quality assurance.  This level of transparency
-                allows researchers and practitioners to fully
-                understand the dataset’s composition and make
-                informed decisions when using TxT360 for training.
-                Additionally, TxT360 includes detailed
-                documentation and analysis of the data, covering
-                distribution statistics, domain coverage, and
-                processing pipeline, which helps users navigate and
-                utilize the dataset effectively.  Overall, TxT360
-                represents a significant step forward in the
-                availability and transparency of large-scale
-                training data for language models, setting a new
-                standard for dataset quality and openness.""")
+intro_list = P(
+    "We present TxT360, the Trillion eXtracted Text corpus, a 5.7T token dataset for pretraining projects that:"
+)
 
-previous_background = P(
-                """ The quality and size of a pre-training dataset
-                    play a crucial role in the performance of large
-                    language models (LLMs). The community has
-                    introduced a variety of datasets for this purpose,
-                    including purely web-based datasets like RefinedWeb
-                    [1], RedPajama-Data-V2 [2], DCLM [3], and
-                    FineWeb [4], as well as comprehensive datasets
-                    derived from multiple highly-curated data sources
-                    such as The Pile [5], RedPajama-Data-V1 [6], and
-                    Dolma [7] . It is commonly known that web-based
-                    datasets provide a vast quantity of data, while
-                    highly-curated multi-source datasets consistently
-                    deliver high quality and diversity, both critical
-                    for effective LLM pre-training.  However, despite
-                    the advancements in both types of data, each type
-                    of dataset has its limitations. For instance, the
-                    processing scripts for the web dataset, RefinedWeb,
-                    known for its high quality, are not public, and
-                    only about 10% of the entire dataset has been
-                    disclosed. Conversely, the web component of
-                    existing highly-curated multi-source datasets is
-                    relatively small compared to purely web-based
-                    datasets, limiting their coverage and diversity
-                    compared to the scale of information from the
-                    internet.  By integrating the extensive reach of
-                    web data with the exceptional quality of curated
-                    sources, TxT360 is crafted to meet and surpass the
-                    rigorous standards required for state-of-the-art
-                    LLM pre-training. """
-            ),
+intro_list1 = Ol(
+    Li(
+        "Curates commonly used pretraining datasets, including all CommonCrawl",
+        style="margin-bottom: 5px",
+    ),
+    Li(
+        "Employs carefully selected filters designed for each data source",
+        style="margin-bottom: 5px",
+    ),
+    Li(
+        "Provides only unique data elements via globally deduplicated across all datasets",
+        style="margin-bottom: 5px",
+    ),
+    Li(
+        "Retains all deduplication metadata for custom upweighting",
+        style="margin-bottom: 5px",
+    ),
+    Li(
+        "Is Production ready! Download here [link to HF repo]",
+        style="margin-bottom: 5px",
+    ),
+)
 
-previous_content = P("""The performance of a large language model (LLM)
-                    depends heavily on the quality and size of its
-                    pretraining dataset. However, the pretraining
-                    datasets for state-of-the-art open LLMs like Llama
-                    3 and Mixtral are not publicly available and very
-                    little is known about how they were created.
-                    Reading time: 45 min. For the best reading
-                    experience, we recommend not using a mobile phone.
-                    Recently, we released 🍷 FineWeb, a new,
-                    large-scale (15-trillion tokens, 44TB disk space)
-                    dataset for LLM pretraining. FineWeb is derived
-                    from 96 CommonCrawl snapshots and produces
-                    better-performing LLMs than other open pretraining
-                    datasets. To bring more clarity in machine learning
-                    and advance the open understanding of how to train
-                    good quality large language models, we carefully
-                    documented and ablated all of the design choices
-                    used in FineWeb, including in-depth investigations
-                    of deduplication and filtering strategies. The
-                    present long form report is a deep dive in how to
-                    create a large and high-quality web-scale dataset
-                    for LLM pretraining. The dataset itself, 🍷
-                    FineWeb, is available here.  We are extremely
-                    thankful to the whole distill.pub team (Christopher
-                    Olah, Shan Carter, Ludwig Schubert in particular)
-                    for creating the template on which we based this
-                    blog post. Thanks also for inspiring us with
-                    exquisitely crafted articles and blog posts.  In
-                    this report we also introduce 📚 FineWeb-Edu, a
-                    subset of FineWeb constructed using scalable
-                    automated high-quality annotations for educational
-                    value, and which outperforms all openly accessible
-                    web-datasets on a number of educational benchmarks
-                    such as MMLU, ARC, and OpenBookQA. 📚 FineWeb-Edu
-                    is available in two sizes/filtering-level: 1.3
-                    trillion (very high educational content) and 5.4
-                    trillion (high educational content) tokens (all
-                    tokens are measured with GPT2 tokenizer). You can
-                    download it here.  Both datasets are released under
-                    the permissive ODC-By 1.0 license TLDR: This blog
-                    covers a discussion on processing and evaluating
-                    data quality at scale, the 🍷 FineWeb recipe
-                    (listing and explaining all of our design choices),
-                    and the process followed to create its 📚
-                    FineWeb-Edu subset."""),
 
-previous_conclusion = P("""This is the conclusion section where we
-                summarize the key points discussed in the blog post
-                and provide final thoughts."""),
-    
 @app.get("/intro")
 def intro():
     return Div(
@@ -274,26 +222,43 @@ def intro():
         ),
         Section(
             H3("Global Deduplication"),
-            P("TxT360 curated a wide range of datasets, including a whopping 99 Common Crawl Dumps and  a list of high quality datasets: StackExchange, Wikipedia, Arxiv, USPTO, DM Math, HackerNews, Ubuntu IRC, Europarl, FreeLaw, PG19, S2ORC, PhilPapers, PubMed Abstracts, and PubMed Central. For the first time in a released dataset, we locally and globally deduplicated the data across each dataset creating the highest quality data available."),
+            P(
+                "TxT360 curated a wide range of datasets, including a whopping 99 Common Crawl Dumps and  a list of high quality datasets: StackExchange, Wikipedia, Arxiv, USPTO, DM Math, HackerNews, Ubuntu IRC, Europarl, FreeLaw, PG19, S2ORC, PhilPapers, PubMed Abstracts, and PubMed Central. For the first time in a released dataset, we locally and globally deduplicated the data across each dataset creating the highest quality data available."
+            ),
             id="section2",
         ),
         Section(
             H3("Controllable Upweighting for Flexible Data Sample Weight Control"),
-            P("In large-scale corpora like CommonCrawl, text duplication is a frequent occurrence. Duplication can be considered as a natural upsampling of some data points. Recent studies have highlighted the potential drawbacks of oversampling specific data points, which can negatively impact pretraining performance [2205.10487]. However, when samples are repeated appropriately, the performance can actually improve [2306.01116, 2305.16264, 2406.11794, FineWeb]. Despite this, there is currently no widely accepted best practice for data sampling, and it’s unlikely that a one-size-fits-all approach will emerge given the scale of these datasets. Previous work either leaves the deduplication process to the user (as seen in RedPajama V2 and DCLM-Pool) or provides a corpus that has been downsampled in a specific manner (such as in FineWeb and RefinedWeb)."),
-            P("Given the high cost of deduplication, TxT360 offers a complete deduplication across all datasets (so you don’t have to). Additionally, TxT360 maintains detailed metadata for each sample, including the frequency and location of duplicates. This metadata gives pretrainers the flexibility to adjust the weight of samples as needed. In principle, one can recover the original dataset distribution (footnote: this approach also means a smaller size on disk). We will demonstrate a simple upsampling strategy that results in an effective pretraining dataset. "),
+            P(
+                "In large-scale corpora like CommonCrawl, text duplication is a frequent occurrence. Duplication can be considered as a natural upsampling of some data points. Recent studies have highlighted the potential drawbacks of oversampling specific data points, which can negatively impact pretraining performance [2205.10487]. However, when samples are repeated appropriately, the performance can actually improve [2306.01116, 2305.16264, 2406.11794, FineWeb]. Despite this, there is currently no widely accepted best practice for data sampling, and it’s unlikely that a one-size-fits-all approach will emerge given the scale of these datasets. Previous work either leaves the deduplication process to the user (as seen in RedPajama V2 and DCLM-Pool) or provides a corpus that has been downsampled in a specific manner (such as in FineWeb",
+                D_cite(bibtex_key="fineweb"),
+                "and RefinedWeb",
+                D_cite(bibtex_key="refinedweb"),
+                ").",
+            ),
+            P(
+                "Given the high cost of deduplication, TxT360 offers a complete deduplication across all datasets (so you don’t have to). Additionally, TxT360 maintains detailed metadata for each sample, including the frequency and location of duplicates. This metadata gives pretrainers the flexibility to adjust the weight of samples as needed. In principle, one can recover the original dataset distribution (footnote: this approach also means a smaller size on disk). We will demonstrate a simple upsampling strategy that results in an effective pretraining dataset. "
+            ),
             id="section3",
         ),
         Section(
             H3("Full and Openly Documented Production Ready Pretraining Corpus"),
-            P("We cover every aspect of the decisions made to produce the dataset, including document selection, filtering, quality assurance, deduplication, standardization and PII.  Our reasoning is thoroughly explained, ensuring transparency and replicability. "),
+            P(
+                "We cover every aspect of the decisions made to produce the dataset, including document selection, filtering, quality assurance, deduplication, standardization and PII.  Our reasoning is thoroughly explained, ensuring transparency and replicability. "
+            ),
             P("Our code is open sourced here[link to github]."),
-            P("The dataset is ready for immediate download directly from Hugging Face [link]."),
-            P("In the remainder of this blog post, we will walk you through the entire process and the rationale behind each decision. Enjoy!"),
+            P(
+                "The dataset is ready for immediate download directly from Hugging Face [link]."
+            ),
+            P(
+                "In the remainder of this blog post, we will walk you through the entire process and the rationale behind each decision. Enjoy!"
+            ),
             id="section4",
         ),
         id="inner-text",
     )
 
+
 rt("/overview")(overview.overview)
 rt("/curated")(curated.curated)
 rt("/curated/{target}")(curated.update)

requirements.txt

@@ -6,4 +6,4 @@ pandas
 Jinja2
 rich
 jsonlines
-bibtexparser
+pybtex