ar:
  name: ar
  pretty_name: Arabic
  n_docs: 50539
  word_count: 8911565
  char_count: 45233908
  sources:
  - filepath: ar/biomedical.jsonl.gz
    name: biomedical
    pretty_name: Biomedical
    n_docs: 15069
    word_count: 979234
    char_count: 5000189
    hf_datasets:
    - premio-ai/TheArabicPile_Medical
  - filepath: ar/books.jsonl.gz
    name: books
    pretty_name: Books
    n_docs: 8063
    word_count: 926754
    char_count: 5024191
    hf_datasets:
    - ImruQays/Rasaif-Classical-Arabic-English-Parallel-texts
  - filepath: ar/lyrics.jsonl.gz
    name: lyrics
    pretty_name: Lyrics
    n_docs: 6703
    word_count: 1040406
    char_count: 5000051
    hf_datasets:
    - premio-ai/TheArabicPile_Lyrics
  - filepath: ar/news.jsonl.gz
    name: news
    pretty_name: News
    n_docs: 3111
    word_count: 890366
    char_count: 5000288
    hf_datasets:
    - AIR-Bench/qa_news_ar
  - filepath: ar/poetry.jsonl.gz
    name: poetry
    pretty_name: Poetry
    n_docs: 4072
    word_count: 1190889
    char_count: 5005196
    hf_datasets:
    - premio-ai/TheArabicPile_Poetry
  - filepath: ar/speech_text.jsonl.gz
    name: speech_text
    pretty_name: Live Speech
    n_docs: 526
    word_count: 982503
    char_count: 5014133
    hf_datasets:
    - bigscience-data/roots_ar_ted_talks_iwslt
  - filepath: ar/subtitles.jsonl.gz
    name: subtitles
    pretty_name: Subtitles
    n_docs: 9811
    word_count: 1090390
    char_count: 5186417
    hf_datasets:
    - badranx/opus_raw
  - filepath: ar/web.jsonl.gz
    name: web
    pretty_name: Web
    n_docs: 1965
    word_count: 907476
    char_count: 5001915
    hf_datasets:
    - HuggingFaceFW/fineweb-2
  - filepath: ar/wiki.jsonl.gz
    name: wiki
    pretty_name: Wikipedia
    n_docs: 1219
    word_count: 903547
    char_count: 5001528
    hf_datasets:
    - bigscience-data/roots_ar_wikipedia
  parities:
  - filepath: ar/parity.jsonl.gz
    name: parity_en
    pretty_name: Parity (English)
    n_docs: 2009
    word_count: 41693
    char_count: 226881
    reference_filepath: en/parity.jsonl.gz
    reference_lang: en
    reference_lang_pretty_name: English
    reference_word_count: 48421
    reference_char_count: 257160
    hf_datasets:
    - openlanguagedata/flores_plus
code:
  name: code
  pretty_name: Programming Languages
  n_docs: 19820
  word_count: 18976907
  char_count: 90121858
  sources:
  - filepath: code/c++.jsonl.gz
    name: c++
    pretty_name: C++
    n_docs: 654
    word_count: 1080031
    char_count: 5016576
    hf_datasets:
    - bigcode/the-stack-smol
  - filepath: code/c.jsonl.gz
    name: c
    pretty_name: C
    n_docs: 806
    word_count: 1067337
    char_count: 5011904
    hf_datasets:
    - bigcode/the-stack-smol
  - filepath: code/css.jsonl.gz
    name: css
    pretty_name: CSS
    n_docs: 680
    word_count: 1173487
    char_count: 5012974
    hf_datasets:
    - bigcode/the-stack-smol
  - filepath: code/dockerfile.jsonl.gz
    name: dockerfile
    pretty_name: Dockerfile
    n_docs: 2277
    word_count: 824868
    char_count: 5001545
    hf_datasets:
    - bigcode/the-stack-smol
  - filepath: code/go.jsonl.gz
    name: go
    pretty_name: Go
    n_docs: 796
    word_count: 1118275
    char_count: 5009452
    hf_datasets:
    - bigcode/the-stack-smol
  - filepath: code/html.jsonl.gz
    name: html
    pretty_name: HTML
    n_docs: 499
    word_count: 1282806
    char_count: 5007475
    hf_datasets:
    - bigcode/the-stack-smol
  - filepath: code/java.jsonl.gz
    name: java
    pretty_name: Java
    n_docs: 1049
    word_count: 872720
    char_count: 5005107
    hf_datasets:
    - bigcode/the-stack-smol
  - filepath: code/javascript.jsonl.gz
    name: javascript
    pretty_name: JavaScript
    n_docs: 948
    word_count: 1192196
    char_count: 5000281
    hf_datasets:
    - bigcode/the-stack-smol
  - filepath: code/json.jsonl.gz
    name: json
    pretty_name: JSON
    n_docs: 3019
    word_count: 1024851
    char_count: 5000156
    hf_datasets:
    - bigcode/starcoderdata
  - filepath: code/lua.jsonl.gz
    name: lua
    pretty_name: Lua
    n_docs: 862
    word_count: 1154199
    char_count: 5007613
    hf_datasets:
    - bigcode/the-stack-smol
  - filepath: code/markdown.jsonl.gz
    name: markdown
    pretty_name: Markdown
    n_docs: 1025
    word_count: 930674
    char_count: 5001675
    hf_datasets:
    - bigcode/the-stack-smol
  - filepath: code/php.jsonl.gz
    name: php
    pretty_name: PHP
    n_docs: 1209
    word_count: 688741
    char_count: 5004026
    hf_datasets:
    - bigcode/the-stack-smol
  - filepath: code/python.jsonl.gz
    name: python
    pretty_name: Python
    n_docs: 805
    word_count: 1012566
    char_count: 5002609
    hf_datasets:
    - bigcode/the-stack-smol
  - filepath: code/rust.jsonl.gz
    name: rust
    pretty_name: Rust
    n_docs: 662
    word_count: 1087446
    char_count: 5016551
    hf_datasets:
    - bigcode/the-stack-smol
  - filepath: code/shell.jsonl.gz
    name: shell
    pretty_name: Shell
    n_docs: 1750
    word_count: 1029915
    char_count: 5000436
    hf_datasets:
    - bigcode/the-stack-smol
  - filepath: code/sql.jsonl.gz
    name: sql
    pretty_name: SQL
    n_docs: 837
    word_count: 1202794
    char_count: 5000476
    hf_datasets:
    - bigcode/the-stack-smol
  - filepath: code/tex.jsonl.gz
    name: tex
    pretty_name: TeX
    n_docs: 797
    word_count: 1128283
    char_count: 5007507
    hf_datasets:
    - bigcode/the-stack-smol
  - filepath: code/typescript.jsonl.gz
    name: typescript
    pretty_name: TypeScript
    n_docs: 1145
    word_count: 1105718
    char_count: 5015495
    hf_datasets:
    - bigcode/the-stack-smol
  parities: null
de:
  name: de
  pretty_name: German
  n_docs: 19740
  word_count: 5876322
  char_count: 35215590
  sources:
  - filepath: de/biomedical.jsonl.gz
    name: biomedical
    pretty_name: Biomedical
    n_docs: 4740
    word_count: 735182
    char_count: 5000777
    hf_datasets:
    - rntc/pubmed_preprocess
  - filepath: de/legal.jsonl.gz
    name: legal
    pretty_name: Legal
    n_docs: 334
    word_count: 783409
    char_count: 5011949
    hf_datasets:
    - joelniklaus/Multi_Legal_Pile
  - filepath: de/news.jsonl.gz
    name: news
    pretty_name: News
    n_docs: 1998
    word_count: 818342
    char_count: 5001257
    hf_datasets:
    - AIR-Bench/qa_news_de
  - filepath: de/papers.jsonl.gz
    name: papers
    pretty_name: Scientific Papers
    n_docs: 58
    word_count: 814548
    char_count: 5035408
    hf_datasets:
    - almanach/HALvest
  - filepath: de/subtitles.jsonl.gz
    name: subtitles
    pretty_name: Subtitles
    n_docs: 9439
    word_count: 1117069
    char_count: 5151049
    hf_datasets:
    - badranx/opus_raw
  - filepath: de/web.jsonl.gz
    name: web
    pretty_name: Web
    n_docs: 1732
    word_count: 810044
    char_count: 5014523
    hf_datasets:
    - HuggingFaceFW/fineweb-2
  - filepath: de/wiki.jsonl.gz
    name: wiki
    pretty_name: Wikipedia
    n_docs: 1439
    word_count: 797728
    char_count: 5000627
    hf_datasets:
    - wikimedia/wikipedia
  parities:
  - filepath: de/parity.jsonl.gz
    name: parity_en
    pretty_name: Parity (English)
    n_docs: 2009
    word_count: 49054
    char_count: 300367
    reference_filepath: en/parity.jsonl.gz
    reference_lang: en
    reference_lang_pretty_name: English
    reference_word_count: 48421
    reference_char_count: 257160
    hf_datasets:
    - openlanguagedata/flores_plus
en:
  name: en
  pretty_name: English
  n_docs: 38155
  word_count: 12361906
  char_count: 61999845
  sources:
  - filepath: en/biomedical.jsonl.gz
    name: biomedical
    pretty_name: Biomedical
    n_docs: 4545
    word_count: 896969
    char_count: 5004335
    hf_datasets:
    - rntc/pubmed_preprocess
  - filepath: en/books.jsonl.gz
    name: books
    pretty_name: Books
    n_docs: 257
    word_count: 1021530
    char_count: 5014929
    hf_datasets:
    - orionweller/books_mds_incremental
  - filepath: en/emoji.jsonl.gz
    name: emoji
    pretty_name: Emoji Heavy
    n_docs: 9952
    word_count: 1523263
    char_count: 6799588
    hf_datasets:
    - Orion-zhen/dpo-mathinstuct-emoji
    - Orion-zhen/dpo-codealpaca-emoji
    - Orion-zhen/dpo-physical-reasoning-emoji
    - ElKulako/stocktwits-emoji
  - filepath: en/finance.jsonl.gz
    name: finance
    pretty_name: Finance
    n_docs: 3676
    word_count: 862599
    char_count: 5001747
    hf_datasets:
    - AIR-Bench/qa_finance_en
  - filepath: en/legal.jsonl.gz
    name: legal
    pretty_name: Legal
    n_docs: 468
    word_count: 808258
    char_count: 5011279
    hf_datasets:
    - joelniklaus/Multi_Legal_Pile
  - filepath: en/news.jsonl.gz
    name: news
    pretty_name: News
    n_docs: 1654
    word_count: 969568
    char_count: 5000743
    hf_datasets:
    - orionweller/cc_news_mds_incremental
  - filepath: en/papers.jsonl.gz
    name: papers
    pretty_name: Scientific Papers
    n_docs: 273
    word_count: 1036211
    char_count: 5015389
    hf_datasets:
    - orionweller/pes2o_mds_incremental
    - orionweller/arxiv_mds_incremental
  - filepath: en/social.jsonl.gz
    name: social
    pretty_name: Social Networks
    n_docs: 4882
    word_count: 1049919
    char_count: 5000428
    hf_datasets:
    - orionweller/reddit_mds_incremental
  - filepath: en/speech_text.jsonl.gz
    name: speech_text
    pretty_name: Live Speech
    n_docs: 421
    word_count: 1064613
    char_count: 5014876
    hf_datasets:
    - bigscience-data/roots_en_ted_talks_iwslt
  - filepath: en/subtitles.jsonl.gz
    name: subtitles
    pretty_name: Subtitles
    n_docs: 9450
    word_count: 1189223
    char_count: 5132323
    hf_datasets:
    - badranx/opus_raw
  - filepath: en/web.jsonl.gz
    name: web
    pretty_name: Web
    n_docs: 1589
    word_count: 976567
    char_count: 5000934
    hf_datasets:
    - HuggingFaceFW/fineweb
  - filepath: en/wiki.jsonl.gz
    name: wiki
    pretty_name: Wikipedia
    n_docs: 988
    word_count: 963186
    char_count: 5003274
    hf_datasets:
    - orionweller/wikipedia_mds_incremental
  parities: null
es:
  name: es
  pretty_name: Spanish
  n_docs: 25315
  word_count: 8420656
  char_count: 45168212
  sources:
  - filepath: es/biomedical.jsonl.gz
    name: biomedical
    pretty_name: Biomedical
    n_docs: 7583
    word_count: 892676
    char_count: 5000619
    hf_datasets:
    - rntc/pubmed_preprocess
  - filepath: es/books.jsonl.gz
    name: books
    pretty_name: Books
    n_docs: 251
    word_count: 970246
    char_count: 5012326
    hf_datasets:
    - jorgeortizfuentes/spanish_books
  - filepath: es/legal.jsonl.gz
    name: legal
    pretty_name: Legal
    n_docs: 420
    word_count: 803517
    char_count: 5005141
    hf_datasets:
    - joelniklaus/Multi_Legal_Pile
  - filepath: es/news.jsonl.gz
    name: news
    pretty_name: News
    n_docs: 2313
    word_count: 933743
    char_count: 5002734
    hf_datasets:
    - LeoCordoba/CC-NEWS-ES
  - filepath: es/papers.jsonl.gz
    name: papers
    pretty_name: Scientific Papers
    n_docs: 2172
    word_count: 879204
    char_count: 5006997
    hf_datasets:
    - eduagarcia/scielo_abstracts
    - almanach/HALvest
  - filepath: es/speech_text.jsonl.gz
    name: speech_text
    pretty_name: Live Speech
    n_docs: 425
    word_count: 984288
    char_count: 5009865
    hf_datasets:
    - bigscience-data/roots_es_ted_talks_iwslt
  - filepath: es/subtitles.jsonl.gz
    name: subtitles
    pretty_name: Subtitles
    n_docs: 9463
    word_count: 1101024
    char_count: 5123037
    hf_datasets:
    - badranx/opus_raw
  - filepath: es/web.jsonl.gz
    name: web
    pretty_name: Web
    n_docs: 1453
    word_count: 930615
    char_count: 5001997
    hf_datasets:
    - HuggingFaceFW/fineweb-2
  - filepath: es/wiki.jsonl.gz
    name: wiki
    pretty_name: Wikipedia
    n_docs: 1235
    word_count: 925343
    char_count: 5005496
    hf_datasets:
    - bigscience-data/roots_es_wikipedia
  parities:
  - filepath: es/parity.jsonl.gz
    name: parity_en
    pretty_name: Parity (English)
    n_docs: 2009
    word_count: 57126
    char_count: 306170
    reference_filepath: en/parity.jsonl.gz
    reference_lang: en
    reference_lang_pretty_name: English
    reference_word_count: 48421
    reference_char_count: 257160
    hf_datasets:
    - openlanguagedata/flores_plus
fr:
  name: fr
  pretty_name: French
  n_docs: 26934
  word_count: 8594454
  char_count: 45185903
  sources:
  - filepath: fr/biomedical.jsonl.gz
    name: biomedical
    pretty_name: Biomedical
    n_docs: 3957
    word_count: 884878
    char_count: 5001744
    hf_datasets:
    - rntc/pubmed_preprocess
  - filepath: fr/finance.jsonl.gz
    name: finance
    pretty_name: Finance
    n_docs: 6156
    word_count: 925634
    char_count: 5000235
    hf_datasets:
    - AIR-Bench/qa_finance_fr
  - filepath: fr/legal.jsonl.gz
    name: legal
    pretty_name: Legal
    n_docs: 399
    word_count: 966888
    char_count: 5017675
    hf_datasets:
    - joelniklaus/Multi_Legal_Pile
  - filepath: fr/news.jsonl.gz
    name: news
    pretty_name: News
    n_docs: 3045
    word_count: 952717
    char_count: 5001112
    hf_datasets:
    - kamarko/ccnews-french-subset
    - AIR-Bench/qa_news_fr
  - filepath: fr/papers.jsonl.gz
    name: papers
    pretty_name: Scientific Papers
    n_docs: 274
    word_count: 890713
    char_count: 5010625
    hf_datasets:
    - almanach/HALvest
  - filepath: fr/speech_text.jsonl.gz
    name: speech_text
    pretty_name: Live Speech
    n_docs: 401
    word_count: 975190
    char_count: 5006652
    hf_datasets:
    - bigscience-data/roots_fr_ted_talks_iwslt
  - filepath: fr/subtitles.jsonl.gz
    name: subtitles
    pretty_name: Subtitles
    n_docs: 9684
    word_count: 1129112
    char_count: 5145257
    hf_datasets:
    - badranx/opus_raw
  - filepath: fr/web.jsonl.gz
    name: web
    pretty_name: Web
    n_docs: 1661
    word_count: 948022
    char_count: 5001436
    hf_datasets:
    - HuggingFaceFW/fineweb-2
  - filepath: fr/wiki.jsonl.gz
    name: wiki
    pretty_name: Wikipedia
    n_docs: 1357
    word_count: 921300
    char_count: 5001167
    hf_datasets:
    - bigscience-data/roots_fr_wikipedia
  parities:
  - filepath: fr/parity.jsonl.gz
    name: parity_en
    pretty_name: Parity (English)
    n_docs: 2009
    word_count: 56532
    char_count: 306427
    reference_filepath: en/parity.jsonl.gz
    reference_lang: en
    reference_lang_pretty_name: English
    reference_word_count: 48421
    reference_char_count: 257160
    hf_datasets:
    - openlanguagedata/flores_plus
it:
  name: it
  pretty_name: Italian
  n_docs: 29414
  word_count: 5462293
  char_count: 30160296
  sources:
  - filepath: it/biomedical.jsonl.gz
    name: biomedical
    pretty_name: Biomedical
    n_docs: 15993
    word_count: 846729
    char_count: 5000031
    hf_datasets:
    - rntc/pubmed_preprocess
    - HiTZ/Multilingual-Medical-Corpus
  - filepath: it/legal.jsonl.gz
    name: legal
    pretty_name: Legal
    n_docs: 421
    word_count: 863903
    char_count: 5005542
    hf_datasets:
    - joelniklaus/Multi_Legal_Pile
  - filepath: it/papers.jsonl.gz
    name: papers
    pretty_name: Scientific Papers
    n_docs: 281
    word_count: 883433
    char_count: 5002831
    hf_datasets:
    - almanach/HALvest
  - filepath: it/subtitles.jsonl.gz
    name: subtitles
    pretty_name: Subtitles
    n_docs: 9914
    word_count: 1102128
    char_count: 5138792
    hf_datasets:
    - badranx/opus_raw
  - filepath: it/web.jsonl.gz
    name: web
    pretty_name: Web
    n_docs: 1281
    word_count: 887675
    char_count: 5010872
    hf_datasets:
    - HuggingFaceFW/fineweb-2
  - filepath: it/wiki.jsonl.gz
    name: wiki
    pretty_name: Wikipedia
    n_docs: 1524
    word_count: 878425
    char_count: 5002228
    hf_datasets:
    - wikimedia/wikipedia
  parities:
  - filepath: it/parity.jsonl.gz
    name: parity_en
    pretty_name: Parity (English)
    n_docs: 2009
    word_count: 53198
    char_count: 303228
    reference_filepath: en/parity.jsonl.gz
    reference_lang: en
    reference_lang_pretty_name: English
    reference_word_count: 48421
    reference_char_count: 257160
    hf_datasets:
    - openlanguagedata/flores_plus
ja:
  name: ja
  pretty_name: Japanese
  n_docs: 35950
  word_count: 19903904
  char_count: 35945203
  sources:
  - filepath: ja/books.jsonl.gz
    name: books
    pretty_name: Books
    n_docs: 674
    word_count: 3274936
    char_count: 5013196
    hf_datasets:
    - globis-university/aozorabunko-clean
  - filepath: ja/haiku.jsonl.gz
    name: haiku
    pretty_name: Haiku
    n_docs: 9290
    word_count: 311295
    char_count: 523737
    hf_datasets:
    - p1atdev/modern_haiku
  - filepath: ja/legal.jsonl.gz
    name: legal
    pretty_name: Legal
    n_docs: 8979
    word_count: 3257782
    char_count: 5027054
    hf_datasets:
    - geniacllm/hanrei
  - filepath: ja/news.jsonl.gz
    name: news
    pretty_name: News
    n_docs: 730
    word_count: 2611419
    char_count: 5001307
    hf_datasets:
    - kajuma/CC-news-2024-July-October-cleaned
  - filepath: ja/papers.jsonl.gz
    name: papers
    pretty_name: Scientific Papers
    n_docs: 440
    word_count: 2177365
    char_count: 5002167
    hf_datasets:
    - kunishou/J-ResearchCorpus
  - filepath: ja/subtitles.jsonl.gz
    name: subtitles
    pretty_name: Subtitles
    n_docs: 9893
    word_count: 2889701
    char_count: 5375927
    hf_datasets:
    - badranx/opus_raw
  - filepath: ja/web.jsonl.gz
    name: web
    pretty_name: Web
    n_docs: 3507
    word_count: 2760359
    char_count: 5000434
    hf_datasets:
    - HuggingFaceFW/fineweb-2
  - filepath: ja/wiki.jsonl.gz
    name: wiki
    pretty_name: Wikipedia
    n_docs: 2437
    word_count: 2621047
    char_count: 5001381
    hf_datasets:
    - wikimedia/wikipedia
  parities:
  - filepath: ja/parity.jsonl.gz
    name: parity_en
    pretty_name: Parity (English)
    n_docs: 2009
    word_count: 67134
    char_count: 111601
    reference_filepath: en/parity.jsonl.gz
    reference_lang: en
    reference_lang_pretty_name: English
    reference_word_count: 48421
    reference_char_count: 257160
    hf_datasets:
    - openlanguagedata/flores_plus
ko:
  name: ko
  pretty_name: Korean
  n_docs: 28795
  word_count: 8362948
  char_count: 30376957
  sources:
  - filepath: ko/finance.jsonl.gz
    name: finance
    pretty_name: Finance
    n_docs: 7684
    word_count: 1383653
    char_count: 5015406
    hf_datasets:
    - BCCard/BCCard-Finance-Kor-QnA
  - filepath: ko/legal.jsonl.gz
    name: legal
    pretty_name: Legal
    n_docs: 2891
    word_count: 1345744
    char_count: 5018651
    hf_datasets:
    - joonhok-exo-ai/korean_law_open_data_precedents
    - jihye-moon/klac_legal_aid_counseling
  - filepath: ko/news.jsonl.gz
    name: news
    pretty_name: News
    n_docs: 4067
    word_count: 1340591
    char_count: 5001633
    hf_datasets:
    - AIR-Bench/qa_news_ko
    - daekeun-ml/naver-news-summarization-ko
  - filepath: ko/subtitles.jsonl.gz
    name: subtitles
    pretty_name: Subtitles
    n_docs: 9956
    word_count: 1563734
    char_count: 5338498
    hf_datasets:
    - badranx/opus_raw
  - filepath: ko/web.jsonl.gz
    name: web
    pretty_name: Web
    n_docs: 2341
    word_count: 1368695
    char_count: 5001241
    hf_datasets:
    - HuggingFaceFW/fineweb-2
  - filepath: ko/wiki.jsonl.gz
    name: wiki
    pretty_name: Wikipedia
    n_docs: 1856
    word_count: 1360531
    char_count: 5001528
    hf_datasets:
    - wikimedia/wikipedia
  parities:
  - filepath: ko/parity.jsonl.gz
    name: parity_en
    pretty_name: Parity (English)
    n_docs: 2009
    word_count: 36250
    char_count: 128976
    reference_filepath: en/parity.jsonl.gz
    reference_lang: en
    reference_lang_pretty_name: English
    reference_word_count: 48421
    reference_char_count: 257160
    hf_datasets:
    - openlanguagedata/flores_plus
pl:
  name: pl
  pretty_name: Polish
  n_docs: 15763
  word_count: 4131439
  char_count: 25168261
  sources:
  - filepath: pl/legal.jsonl.gz
    name: legal
    pretty_name: Legal
    n_docs: 544
    word_count: 675424
    char_count: 5006421
    hf_datasets:
    - joelniklaus/Multi_Legal_Pile
  - filepath: pl/news.jsonl.gz
    name: news
    pretty_name: News
    n_docs: 2601
    word_count: 775307
    char_count: 5002359
    hf_datasets:
    - WiktorS/polish-news
    - winterkitsune/elka-pl-news
  - filepath: pl/subtitles.jsonl.gz
    name: subtitles
    pretty_name: Subtitles
    n_docs: 9758
    word_count: 1053609
    char_count: 5156250
    hf_datasets:
    - badranx/opus_raw
  - filepath: pl/web.jsonl.gz
    name: web
    pretty_name: Web
    n_docs: 1532
    word_count: 815487
    char_count: 5000817
    hf_datasets:
    - HuggingFaceFW/fineweb-2
  - filepath: pl/wiki.jsonl.gz
    name: wiki
    pretty_name: Wikipedia
    n_docs: 1328
    word_count: 811612
    char_count: 5002414
    hf_datasets:
    - wikimedia/wikipedia
  parities:
  - filepath: pl/parity.jsonl.gz
    name: parity_en
    pretty_name: Parity (English)
    n_docs: 2009
    word_count: 44308
    char_count: 272953
    reference_filepath: en/parity.jsonl.gz
    reference_lang: en
    reference_lang_pretty_name: English
    reference_word_count: 48421
    reference_char_count: 257160
    hf_datasets:
    - openlanguagedata/flores_plus
pt:
  name: pt
  pretty_name: Portuguese
  n_docs: 30528
  word_count: 9603267
  char_count: 50195889
  sources:
  - filepath: pt/biomedical.jsonl.gz
    name: biomedical
    pretty_name: Biomedical
    n_docs: 5949
    word_count: 871754
    char_count: 5000779
    hf_datasets:
    - rntc/pubmed_preprocess
  - filepath: pt/books.jsonl.gz
    name: books
    pretty_name: Books
    n_docs: 292
    word_count: 946141
    char_count: 5013054
    hf_datasets:
    - eduagarcia/pd_books_pt
  - filepath: pt/legal.jsonl.gz
    name: legal
    pretty_name: Legal
    n_docs: 1004
    word_count: 920230
    char_count: 5001682
    hf_datasets:
    - eduagarcia/LegalPT_dedup
  - filepath: pt/news.jsonl.gz
    name: news
    pretty_name: News
    n_docs: 2088
    word_count: 926376
    char_count: 5000953
    hf_datasets:
    - eduagarcia/cc_news_pt_v2
  - filepath: pt/papers.jsonl.gz
    name: papers
    pretty_name: Scientific Papers
    n_docs: 280
    word_count: 903768
    char_count: 5008389
    hf_datasets:
    - eduagarcia/scielo_papers
  - filepath: pt/social.jsonl.gz
    name: social
    pretty_name: Social Networks
    n_docs: 7244
    word_count: 1034347
    char_count: 5006373
    hf_datasets:
    - carolina-c4ai/corpus-carolina
    - tallesl/quinquilharia
  - filepath: pt/speech_text.jsonl.gz
    name: speech_text
    pretty_name: Live Speech
    n_docs: 457
    word_count: 990929
    char_count: 5005175
    hf_datasets:
    - bigscience-data/roots_pt_ted_talks_iwslt
  - filepath: pt/subtitles.jsonl.gz
    name: subtitles
    pretty_name: Subtitles
    n_docs: 9874
    word_count: 1151715
    char_count: 5157992
    hf_datasets:
    - badranx/opus_raw
    - badranx/opus_raw
  - filepath: pt/web.jsonl.gz
    name: web
    pretty_name: Web
    n_docs: 1713
    word_count: 924538
    char_count: 5000068
    hf_datasets:
    - HuggingFaceFW/fineweb-2
  - filepath: pt/wiki.jsonl.gz
    name: wiki
    pretty_name: Wikipedia
    n_docs: 1627
    word_count: 933469
    char_count: 5001424
    hf_datasets:
    - bigscience-data/roots_pt_wikipedia
  parities:
  - filepath: pt/parity.jsonl.gz
    name: parity_en
    pretty_name: Parity (English)
    n_docs: 2009
    word_count: 51699
    char_count: 279427
    reference_filepath: en/parity.jsonl.gz
    reference_lang: en
    reference_lang_pretty_name: English
    reference_word_count: 48421
    reference_char_count: 257160
    hf_datasets:
    - openlanguagedata/flores_plus
ru:
  name: ru
  pretty_name: Russian
  n_docs: 24329
  word_count: 5250785
  char_count: 30360746
  sources:
  - filepath: ru/biomedical.jsonl.gz
    name: biomedical
    pretty_name: Biomedical
    n_docs: 6762
    word_count: 844918
    char_count: 5000632
    hf_datasets:
    - rntc/pubmed_preprocess
    - blinoff/medical_qa_ru_data
  - filepath: ru/news.jsonl.gz
    name: news
    pretty_name: News
    n_docs: 2806
    word_count: 818470
    char_count: 5003757
    hf_datasets:
    - AIR-Bench/qa_news_ru
    - IlyaGusev/ru_news
  - filepath: ru/papers.jsonl.gz
    name: papers
    pretty_name: Scientific Papers
    n_docs: 2232
    word_count: 802711
    char_count: 5180509
    hf_datasets:
    - almanach/HALvest
    - mlsa-iai-msu-lab/ru_sci_bench
  - filepath: ru/subtitles.jsonl.gz
    name: subtitles
    pretty_name: Subtitles
    n_docs: 9734
    word_count: 1136577
    char_count: 5165530
    hf_datasets:
    - badranx/opus_raw
  - filepath: ru/web.jsonl.gz
    name: web
    pretty_name: Web
    n_docs: 972
    word_count: 819334
    char_count: 5009522
    hf_datasets:
    - HuggingFaceFW/fineweb-2
  - filepath: ru/wiki.jsonl.gz
    name: wiki
    pretty_name: Wikipedia
    n_docs: 1823
    word_count: 828775
    char_count: 5000796
    hf_datasets:
    - wikimedia/wikipedia
  parities:
  - filepath: ru/parity.jsonl.gz
    name: parity_en
    pretty_name: Parity (English)
    n_docs: 2009
    word_count: 45630
    char_count: 280198
    reference_filepath: en/parity.jsonl.gz
    reference_lang: en
    reference_lang_pretty_name: English
    reference_word_count: 48421
    reference_char_count: 257160
    hf_datasets:
    - openlanguagedata/flores_plus
zh:
  name: zh
  pretty_name: Chinese
  n_docs: 47627
  word_count: 24276657
  char_count: 43209186
  sources:
  - filepath: zh/biomedical.jsonl.gz
    name: biomedical
    pretty_name: Biomedical
    n_docs: 8072
    word_count: 2931192
    char_count: 5000287
    hf_datasets:
    - rntc/pubmed_preprocess
    - DavidLanz/medical_pretrain
  - filepath: zh/emoji.jsonl.gz
    name: emoji
    pretty_name: Emoji Heavy
    n_docs: 8567
    word_count: 1615517
    char_count: 2780513
    hf_datasets:
    - Orion-zhen/dpo-emoji-zh
    - Orion-zhen/dpo-ruozhiba-emoji
    - shareAI/DPO-zh-en-emoji
  - filepath: zh/finance.jsonl.gz
    name: finance
    pretty_name: Finance
    n_docs: 1065
    word_count: 2657651
    char_count: 5002615
    hf_datasets:
    - Duxiaoman-DI/FinCorpus
  - filepath: zh/legal.jsonl.gz
    name: legal
    pretty_name: Legal
    n_docs: 9046
    word_count: 2811261
    char_count: 5027225
    hf_datasets:
    - TigerResearch/tigerbot-law-plugin
  - filepath: zh/news.jsonl.gz
    name: news
    pretty_name: News
    n_docs: 4145
    word_count: 2802470
    char_count: 5001221
    hf_datasets:
    - liswei/news-collection-zhtw
    - AIR-Bench/qa_news_zh
  - filepath: zh/speech_text.jsonl.gz
    name: speech_text
    pretty_name: Live Speech
    n_docs: 1320
    word_count: 3016694
    char_count: 5002517
    hf_datasets:
    - bigscience-data/roots_zh_ted_talks_iwslt
  - filepath: zh/subtitles.jsonl.gz
    name: subtitles
    pretty_name: Subtitles
    n_docs: 9844
    word_count: 2894646
    char_count: 5393731
    hf_datasets:
    - badranx/opus_raw
  - filepath: zh/web.jsonl.gz
    name: web
    pretty_name: Web
    n_docs: 2491
    word_count: 2841988
    char_count: 5000109
    hf_datasets:
    - HuggingFaceFW/fineweb-2
  - filepath: zh/wiki.jsonl.gz
    name: wiki
    pretty_name: Wikipedia
    n_docs: 3077
    word_count: 2705238
    char_count: 5000968
    hf_datasets:
    - bigscience-data/roots_zh-cn_wikipedia
  parities:
  - filepath: zh/parity.jsonl.gz
    name: parity_en
    pretty_name: Parity (English)
    n_docs: 2009
    word_count: 48362
    char_count: 85427
    reference_filepath: en/parity.jsonl.gz
    reference_lang: en
    reference_lang_pretty_name: English
    reference_word_count: 48421
    reference_char_count: 257160
    hf_datasets:
    - openlanguagedata/flores_plus