File size: 77,876 Bytes
630cacb 5137c6e 630cacb 31cd62a 630cacb 31cd62a 630cacb 31cd62a 630cacb 31cd62a 630cacb 31cd62a 630cacb 4fb7f42 630cacb a3bd347 931d5ff a3bd347 6e31ec7 a3bd347 6e31ec7 a3bd347 6e31ec7 a3bd347 6e31ec7 a3bd347 6e31ec7 a3bd347 931d5ff a3bd347 630cacb 4fb7f42 630cacb cca9f44 630cacb 1e61e7a 630cacb 1e61e7a 630cacb 78bb285 630cacb |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 |
---
license: apache-2.0
library_name: transformers
pipeline_tag: text-generation
language:
- bg
- ca
- code
- cs
- cy
- da
- de
- el
- en
- es
- et
- eu
- fi
- fr
- ga
- gl
- hr
- hu
- it
- lt
- lv
- mt
- nl
- nn
- no
- oc
- pl
- pt
- ro
- ru
- sh
- sk
- sl
- sr
- sv
- uk
---
# Salamandra Model Card
Salamandra comes in three different sizes — 2B, 7B and 40B parameters — with their respective base and instruction-tuned variants.
This model card corresponds to the 7B instructed version.
To visit the model cards of other Salamandra versions, please refer to the [Model Index](#model-index).
The entire Salamandra family is released under a permissive [Apache 2.0 license]((https://www.apache.org/licenses/LICENSE-2.0)), allowing both research and commercial use.
Along with the open weights, all training scripts and configuration files are made publicly available in [this GitHub repository](https://github.com/projecte-aina/salamandra).
> [!WARNING]
> **DISCLAIMER:** This model is a first proof-of-concept designed to demonstrate the instruction-following capabilities of recently released base models.
> It has been optimized to engage in conversation but has *NOT* been aligned through RLHF to filter or avoid sensitive topics.
> As a result, it may generate harmful or inappropriate content.
> The team is actively working to enhance its performance through further instruction and alignment with RL techniques.
---
## Model Details
### Description
Transformer-based decoder-only language model that has been pre-trained on 7.8 trillion tokens of highly curated data.
The pre-training corpus contains text in 35 European languages and code.
### Hyperparameters
The full list of hyperparameters for each model can be found [here](https://github.com/projecte-aina/salamandra/tree/main/configs).
### Architecture
| | |
|-------------------------|:--------------|
| Total Parameters | 7,768,117,248 |
| Embedding Parameters | 1,048,576,000 |
| Layers | 32 |
| Hidden size | 4,096 |
| Attention heads | 32 |
| Context length | 8,192 |
| Vocabulary size | 256,000 |
| Precision | bfloat16 |
| Embedding type | RoPE |
| Activation Function | SwiGLU |
| Layer normalization | RMS Norm |
| Flash attention | ✅ |
| Grouped Query Attention | ✅ |
| Num. query groups | 8 |
---
## Intended Use
### Direct Use
The models are intended for both research and commercial use in any of the languages included in the training data.
The base models are intended either for language generation or to be further fine-tuned for specific use-cases.
The instruction-tuned variants can be used as general-purpose assistants, as long as the user is fully aware of the model’s limitations.
### Out-of-scope Use
The model is not intended for malicious activities, such as harming others or violating human rights.
Any downstream application must comply with current laws and regulations.
Irresponsible usage in production environments without proper risk assessment and mitigation is also discouraged.
---
## Hardware and Software
### Training Framework
Pre-training was conducted using NVIDIA’s [NeMo Framework](https://docs.nvidia.com/nemo-framework/index.html),
which leverages PyTorch Lightning for efficient model training in highly distributed settings.
The instruction-tuned versions were produced with [FastChat](https://github.com/lm-sys/FastChat).
### Compute Infrastructure
All models were trained on [MareNostrum 5](https://www.bsc.es/ca/marenostrum/marenostrum-5), a pre-exascale EuroHPC supercomputer hosted and
operated by Barcelona Supercomputing Center.
The accelerated partition is composed of 1,120 nodes with the following specifications:
- 4x Nvidia Hopper GPUs with 64 HBM2 memory
- 2x Intel Sapphire Rapids 8460Y+ at 2.3Ghz and 32c each (64 cores)
- 4x NDR200 (BW per node 800Gb/s)
- 512 GB of Main memory (DDR5)
- 460GB on NVMe storage
|Model|Nodes|GPUs|
|:---:|:---:|:---:|
|2B|64|256|
|7B|128|512|
|40B|256 / 512|1,024 / 2,048|
---
## How to use
The instruction-following models use the commonly adopted ChatML template:
```jinja
{%- if not date_string is defined %}{%- set date_string = "2024-09-30" %}{%- endif %}{{ "<|im_start|>system\nsystem_message\nToday Date: "+ date_string +"<|im_end|>\n" }}{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}
```
Where `system_message` is used to guide the model during generation and `date_string` can be set to allow the model to respond with the current date.
The exact same chat template should be used for an enhanced conversational experience.
The easiest way to apply it is by using the tokenizer's built-in functions, as shown in the following snippet.
```python
from datetime import datetime
from transformers import AutoTokenizer, AutoModelForCausalLM
import transformers
import torch
model_id = "projecte-aina/salamandra-7b-instruct"
text = "At what temperature does water boil?"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
model_id,
device_map="auto",
torch_dtype=torch.bfloat16
)
message = [ { "role": "user", "content": text } ]
date_string = datetime.today().strftime('%Y-%m-%d')
prompt = tokenizer.apply_chat_template(
message,
tokenize=False,
add_generation_prompt=True,
date_string=date_string
)
inputs = tokenizer.encode(prompt, add_special_tokens=False, return_tensors="pt")
outputs = model.generate(input_ids=inputs.to(model.device), max_new_tokens=200)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```
Using this template, each turn is preceded by a `<|im_start|>` delimiter and the role of the entity
(either `user`, for content supplied by the user, or `assistant` for LLM responses), and finished with the `<|im_end|>` token.
---
## Data
### Pretraining Data
The training corpus consists of 2.4 trillion tokens, including 35 European languages and 92 programming languages. It amounts to a total of 33TB of pre-processed text.
Languages were sampled manually by giving x2 oversampling to Spain's co-official languages (Spanish, Catalan, Galician and Basque), code was undersampled by half,
and the rest of the languages were kept as is, resulting in the following distribution:
This highly multilingual corpus is predominantly composed of data from Colossal OSCAR,
which contributes a significant 66.06% of the total tokens.
Following this, Starcoder provides 11.91%, and Spanish Crawling adds 3.34%.
The next largest sources are French FR at 3.12% and Proof Pile at 1.98%.
Other notable contributions include Macocu, Pile of Law, and Eurlex, each contributing around 1.5% to 1.3%.
These major sources collectively form the bulk of the corpus, ensuring a rich and diverse dataset for training the language model.
The remaining 10% comes from smaller sources in various languages.
Feel free to click the expand button below to see the full list of sources.
<details>
<summary>Data Sources</summary>
| Dataset | Language | Source |
|-------------------------------------------------------------------------|--------------------------------------------------------------------------------------------|-----------------------------------------------------------------------------------------------------------------|
| Parlamint corpus | at, bg, cz, dk, ee, es, es-ga, fi, fr, gb, gr, hr, hu, it, lv, nl, no, pl, pt, rs, se, si | (Erjavec et al., 2021) |
| Crawl of Bulgarian news websites | bg | [Link](http://old.dcl.bas.bg/dataset/Bulgarian_news.7z) |
| Bulgarian National Corpus | bg | [Link](http://old.dcl.bas.bg/dataset/BulNC.7z) |
| Colossal OSCAR 1.0 | bg, ca, cs, cy, da, de, el, en, es, et, eu, fi, fr, ga, gl, hr, hu, it, lt, lv, mt, nl, nn, no, oc, pl, pt, ro, ru, sh, sk, sl, sr, sv, uk | (Brack et al., 2024) |
| Wikimedia dumps | bg, ca, cs, da, de, el, en, es, et, eu, fi, fr, ga, gl, hr, hu, it, lt, lv, mt, nl, nn, no, pl, pt, ro, sh, sk, sl, sr, uk | [Link](https://dumps.wikimedia.org/) |
| OpenSubtitlesv2016 | bg, ca, cs, da, de, el, en, es, et, eu, fi, fr, gl, hr, it, lt, lv, nl, no, pl, pt, ro, sk, sl, sr, sv, uk | (Lison & Tiedemann, 2016) |
| MaCoCu web corpus | bg, ca, el, hr, mt, sl, sr, uk | (Bañón et al., 2022) |
| EurLEX-Resources | bg, cs, da, de, el, en, es, et, fi, fr, ga, hr, hu, it, lt, lv, mt, nl, pl, pt, ro, sk, sl, sv | [Link](https://huggingface.co/datasets/joelniklaus/eurlex_resources) |
| MC4-Legal | bg, cs, da, de, el, en, es, et, fi, fr, ga, hu, it, lt, lv, mt, nl, pl, pt, ro, sk, sl, sv | [Link](https://huggingface.co/datasets/joelito/legal-mc4) |
| CURLICAT Corpus | bg, hr, hu, pl, ro, sk, sl | (Váradi et al., 2022) |
| CATalog | ca | (Palomar-Giner et al., 2024) |
| Spanish Crawling | ca, es, eu, gl | Relevant Spanish websites crawling |
| Starcoder | code | (Li et al., 2023) |
| SYN v9: large corpus of written Czech | cs | (Křen et al., 2021) |
| Welsh-GOV | cy | Crawling from [Link](https://www.llyw.cymru) |
| DaNewsroom | da | (Varab & Schluter, 2020) |
| The Danish Parliament Corpus 2009 - 2017, v1 | da | (Hansen, 2018) |
| Danish GigaWord | da | (Strømberg-Derczynski et al., 2021) |
| DK-CLARIN Reference Corpus of General Danish | da | [Link](https://korpus.dsl.dk/clarin/) |
| Open Legal Data - German court decisions and laws | de | (Ostendorff et al., 2020) |
| DeWaC | de | [Link](https://docs.sslmit.unibo.it/doku.php?id=corpora:dewac) |
| Greek Web Corpus | el | (Outsios et al., 2018) |
| Greek Legal Code | el | (Papaloukas et al., 2021) |
| BIGPATENT | en | (Sharma et al., 2019) |
| peS2o | en | (Soldaini & Lo, 2023) |
| PG-19 | en | (Rae et al., 2019) |
| proof-pile | en | [Link](https://huggingface.co/datasets/hoskinson-center/proof-pile) |
| Auxiliary Mathematics Problems and Solutions (AMPS) dataset | en | (Hendrycks et al., 2021) |
| Pile of Law (selected subsets) | en | (Henderson\* et al., 2022) |
| RedPajama-Data T1 (StackExchange subset) | en | (Computer, 2023) |
| The Pile (PhilPapers subset) | en | (Gao et al., 2021) |
| Spanish Legal Domain Corpora | es | (Gutiérrez-Fandiño et al., 2021) |
| HPLTDatasets v1 - Spanish | es | (de Gibert et al., 2024) |
| Legal | es | Internally generated legal dataset: BOE, BORME, Senado, Congreso, Spanish court orders, DOGC |
| Biomedical | es | Internally generated scientific dataset: Dialnet, Scielo, CSIC, TDX, BSC, UCM |
| Scientific | es | Internally generated scientific dataset: Wikipedia LS, Pubmed, MeSpEn, patents, clinical cases, medical crawler |
| Estonian National Corpus 2021 | et | (Koppel & Kallas, 2022) |
| Estonian Reference Corpus | et | [Link](https://www.cl.ut.ee/korpused/segakorpus/) |
| EusCrawl (filtered: no Wikipedia, no NC-licenses) | eu | (Artetxe et al., 2022) |
| Latxa Corpus v1.1 | eu | (Etxaniz et al., 2024) [Link](https://huggingface.co/datasets/HiTZ/latxa-corpus-v1.1)|
| Yle Finnish News Archive | fi | [Link](http://urn.fi/urn:nbn:fi:lb-2021050401) |
| CaBeRnet: a New French Balanced Reference Corpus | fr | (Popa-Fabre et al., 2020) |
| French Public Domain Newspapers | fr | [Link](https://huggingface.co/datasets/PleIAs/French-PD-Newspapers) |
| French Public Domain Books | fr | [Link](https://huggingface.co/datasets/PleIAs/French-PD-Books) |
| The Gaois bilingual corpus of English-Irish legislation (Irish legislation) | ga | [Link](https://portulanclarin.net/repository/browse/the-gaois-bilingual-corpus-of-english-irish-legislation-processed/daeac17c9e3511ea9b7f02420a000407b83de243dc0b469aab41084386c5b80f/) |
| Irish Universal Dependencies | ga | [Link](https://universaldependencies.org/ga/index.html) |
| CorpusNÓS | gl | (de-Dios-Flores et al., 2024) |
| Croatian web corpus hrWaC 2.1 | hr | (Ljubešić & Klubička, 2014) |
| ITWaC | it | [Link](https://docs.sslmit.unibo.it/doku.php?id=corpora:itwac) |
| Corpus of State-related content from the Latvian Web (Processed) | lv | [Link](https://catalog.elra.info/en-us/repository/browse/ELRA-W0169/) |
| Korpus Malti | mt | (Micallef et al., 2022) |
| SoNaR Corpus NC 1.2 | nl | [Link](https://taalmaterialen.ivdnt.org/download/tstc-sonar-corpus/) |
| Norwegian Colossal Corpus | nn, no | (Kummervold et al., 2021) |
| Occitan Corpus | oc | Provided by [IEA](https://www.institutestudisaranesi.cat/) |
| Polish Parliamentary Corpus / Korpus Dyskursu Parlamentarnego | pl | (Ogrodniczuk, 2018) |
| NKJP-PodkorpusMilionowy-1.2 (National Corpus of Polish) | pl | (Lewandowska-Tomaszczyk et al., 2013) |
| Brazilian Portuguese Web as Corpus | pt | (Wagner Filho et al., 2018) |
| ParlamentoPT | pt | (Rodrigues et al., 2023) |
| MARCELL Romanian legislative subcorpus v2 | ro | [Link](https://elrc-share.eu/reposMARCELL%20Romanian%20legislative%20subcorpus%20v2itory/browse/marcell-romanian-legislative-subcorpus-v2/2da548428b9d11eb9c1a00155d026706ce94a6b59ffc4b0e9fb5cd9cebe6889e/) |
| Korpus slovenských právnych predpisov v1.9 | sk | [Link](https://www.juls.savba.sk/data/marcell/legal-sk-20220322-1.9.ver.xz) |
| od-justice 2.0 | sk | [Link](https://www.juls.savba.sk/data/od-justice/od-justice-2.0.ver.xz) |
| Corpus of academic Slovene KAS 2.0 | sl | (Žagar et al., 2022) |
| slWaC web corpus | sl | (Erjavec et al., 2015) |
| SrpKorSubset (news, legal, academic, conversation, literary) | sr | [Link](http://www.korpus.matf.bg.ac.rs/) |
| The Swedish Culturomics Gigaword Corpus | sv | (Rødven-Eide, 2016) |
| Corpus of laws and legal acts of Ukraine | uk | [Link](https://lang.org.ua/en/corpora/#anchor7) |
<details>
<summary>References</summary>
- Abadji, J., Suárez, P. J. O., Romary, L., & Sagot, B. (2021). Ungoliant: An optimized pipeline for the generation of a very large-scale multilingual web corpus (H. Lüngen, M. Kupietz, P. Bański, A. Barbaresi, S. Clematide, & I. Pisetta, Eds.; pp. 1–9). Leibniz-Institut für Deutsche Sprache. [Link](https://doi.org/10.14618/ids-pub-10468)
- Artetxe, M., Aldabe, I., Agerri, R., Perez-de-Viñaspre, O., & Soroa, A. (2022). Does Corpus Quality Really Matter for Low-Resource Languages?
- Bañón, M., Esplà-Gomis, M., Forcada, M. L., García-Romero, C., Kuzman, T., Ljubešić, N., van Noord, R., Sempere, L. P., Ramírez-Sánchez, G., Rupnik, P., Suchomel, V., Toral, A., van der Werff, T., & Zaragoza, J. (2022). MaCoCu: Massive collection and curation of monolingual and bilingual data: Focus on under-resourced languages. Proceedings of the 23rd Annual Conference of the European Association for Machine Translation, 303–304. [Link](https://aclanthology.org/2022.eamt-1.41)
- Brack, M., Ostendorff, M., Suarez, P. O., Saiz, J. J., Castilla, I. L., Palomar-Giner, J., Shvets, A., Schramowski, P., Rehm, G., Villegas, M., & Kersting, K. (2024). Community OSCAR: A Community Effort for Multilingual Web Data. [Link](https://occiglot.eu/papers/Community_Oscar.pdf)
- Computer, T. (2023). RedPajama: An Open Source Recipe to Reproduce LLaMA training dataset [Computer software]. [Link](https://github.com/togethercomputer/RedPajama-Data)
- de Gibert, O., Nail, G., Arefyev, N., Bañón, M., van der Linde, J., Ji, S., Zaragoza-Bernabeu, J., Aulamo, M., Ramírez-Sánchez, G., Kutuzov, A., Pyysalo, S., Oepen, S., & Tiedemann, J. (2024). A New Massive Multilingual Dataset for High-Performance Language Technologies (arXiv:2403.14009). arXiv. [Link](http://arxiv.org/abs/2403.14009)
- Dodge, J., Sap, M., Marasović, A., Agnew, W., Ilharco, G., Groeneveld, D., Mitchell, M., & Gardner, M. (2021). Documenting Large Webtext Corpora: A Case Study on the Colossal Clean Crawled Corpus. In M.-F. Moens, X. Huang, L. Specia, & S. W. Yih (Eds.), Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing (pp. 1286–1305). Association for Computational Linguistics. [Link](https://doi.org/10.18653/v1/2021.emnlp-main.98)
- Erjavec, T., Ljubešić, N., & Logar, N. (2015). The slWaC corpus of the Slovene web. Informatica (Slovenia), 39, 35–42.
- Erjavec, T., Ogrodniczuk, M., Osenova, P., Ljubešić, N., Simov, K., Grigorova, V., Rudolf, M., Pančur, A., Kopp, M., Barkarson, S., Steingrímsson, S. hór, van der Pol, H., Depoorter, G., de Does, J., Jongejan, B., Haltrup Hansen, D., Navarretta, C., Calzada Pérez, M., de Macedo, L. D., … Rayson, P. (2021). Linguistically annotated multilingual comparable corpora of parliamentary debates ParlaMint.ana 2.1. [Link](http://hdl.handle.net/11356/1431)
- Etxaniz, J., Sainz, O., Perez, N., Aldabe, I., Rigau, G., Agirre, E., Ormazabal, A., Artetxe, M., & Soroa, A. (2024). Latxa: An Open Language Model and Evaluation Suite for Basque. [Link] (https://arxiv.org/abs/2403.20266)
- Gao, L., Biderman, S., Black, S., Golding, L., Hoppe, T., Foster, C., Phang, J., He, H., Thite, A., Nabeshima, N., Presser, S., & Leahy, C. (2021). The Pile: An 800GB Dataset of Diverse Text for Language Modeling. CoRR, abs/2101.00027. [Link](https://arxiv.org/abs/2101.00027)
- Gutiérrez-Fandiño, A., Armengol-Estapé, J., Gonzalez-Agirre, A., & Villegas, M. (2021). Spanish Legalese Language Model and Corpora.
- Hansen, D. H. (2018). The Danish Parliament Corpus 2009—2017, v1. [Link](http://hdl.handle.net/20.500.12115/8)
- Henderson*, P., Krass*, M. S., Zheng, L., Guha, N., Manning, C. D., Jurafsky, D., & Ho, D. E. (2022). Pile of Law: Learning Responsible Data Filtering from the Law and a 256GB Open-Source Legal Dataset. arXiv. [Link](https://arxiv.org/abs/2207.00220)
- Hendrycks, D., Burns, C., Kadavath, S., Arora, A., Basart, S., Tang, E., Song, D., & Steinhardt, J. (2021). Measuring Mathematical Problem Solving With the MATH Dataset. NeurIPS.
- Jansen, T., Tong, Y., Zevallos, V., & Suarez, P. O. (2022). Perplexed by Quality: A Perplexity-based Method for Adult and Harmful Content Detection in Multilingual Heterogeneous Web Data.
- Koppel, K., & Kallas, J. (2022). Eesti keele ühendkorpuste sari 2013–2021: Mahukaim eestikeelsete digitekstide kogu. Eesti Rakenduslingvistika Ühingu Aastaraamat Estonian Papers in Applied Linguistics, 18, 207–228. [Link](https://doi.org/10.5128/erya18.12)
- Křen, M., Cvrček, V., Henyš, J., Hnátková, M., Jelínek, T., Kocek, J., Kováříková, D., Křivan, J., Milička, J., Petkevič, V., Procházka, P., Skoumalová, H., Šindlerová, J., & Škrabal, M. (2021). SYN v9: Large corpus of written Czech. [Link](http://hdl.handle.net/11234/1-4635)
- Kreutzer, J., Caswell, I., Wang, L., Wahab, A., van Esch, D., Ulzii-Orshikh, N., Tapo, A., Subramani, N., Sokolov, A., Sikasote, C., Setyawan, M., Sarin, S., Samb, S., Sagot, B., Rivera, C., Rios, A., Papadimitriou, I., Osei, S., Suarez, P. O., … Adeyemi, M. (2022). Quality at a Glance: An Audit of Web-Crawled Multilingual Datasets. Transactions of the Association for Computational Linguistics, 10, 50–72. [Link](https://doi.org/10.1162/tacl_a_00447)
- Kummervold, P. E., De la Rosa, J., Wetjen, F., & Brygfjeld, S. A. (2021). Operationalizing a National Digital Library: The Case for a Norwegian Transformer Model. In S. Dobnik & L. Øvrelid (Eds.), Proceedings of the 23rd Nordic Conference on Computational Linguistics (NoDaLiDa) (pp. 20–29). Linköping University Electronic Press, Sweden. [Link](https://aclanthology.org/2021.nodalida-main.3)
- Lewandowska-Tomaszczyk, B., Górski, R., Łaziński, M., & Przepiórkowski, A. (2013). The National Corpus of Polish (NKJP). Language use and data analysis. 309–319.
- Li, R., Allal, L. B., Zi, Y., Muennighoff, N., Kocetkov, D., Mou, C., Marone, M., Akiki, C., Li, J., Chim, J., Liu, Q., Zheltonozhskii, E., Zhuo, T. Y., Wang, T., Dehaene, O., Davaadorj, M., Lamy-Poirier, J., Monteiro, J., Shliazhko, O., … Vries, H. de. (2023). StarCoder: May the source be with you!
- Lison, P., & Tiedemann, J. (2016). OpenSubtitles2016: Extracting Large Parallel Corpora from Movie and TV Subtitles. In N. Calzolari, K. Choukri, T. Declerck, S. Goggi, M. Grobelnik, B. Maegaard, J. Mariani, H. Mazo, A. Moreno, J. Odijk, & S. Piperidis (Eds.), Proceedings of the Tenth International Conference on Language Resources and Evaluation (LREC’16) (pp. 923–929). European Language Resources Association (ELRA). [Link](https://aclanthology.org/L16-1147)
- Ljubešić, N., & Klubička, F. (2014). Bs,hr,srWaC - Web Corpora of Bosnian, Croatian and Serbian. In F. Bildhauer & R. Schäfer (Eds.), Proceedings of the 9th Web as Corpus Workshop (WaC-9) (pp. 29–35). Association for Computational Linguistics. [Link](https://doi.org/10.3115/v1/W14-0405)
- Micallef, K., Gatt, A., Tanti, M., van der Plas, L., & Borg, C. (2022). Pre-training Data Quality and Quantity for a Low-Resource Language: New Corpus and BERT Models for Maltese. Proceedings of the Third Workshop on Deep Learning for Low-Resource Natural Language Processing, 90–101. [Link](https://doi.org/10.18653/v1/2022.deeplo-1.10)
- Ogrodniczuk, M. (2018). Polish Parliamentary Corpus. [Link](https://api.semanticscholar.org/CorpusID:235134113)
- Ostendorff, M., Blume, T., & Ostendorff, S. (2020). Towards an Open Platform for Legal Information. Proceedings of the ACM/IEEE Joint Conference on Digital Libraries in 2020, 385–388. [Link](https://doi.org/10.1145/3383583.3398616)
- Ostendorff, M., Suarez, P. O., Lage, L. F., & Rehm, G. (2024). LLM-Datasets: An Open Framework for Pretraining Datasets of Large Language Models. First Conference on Language Modeling. [Link](https://openreview.net/forum?id=5RdIMlGLXL)
- Outsios, S., Skianis, K., Meladianos, P., Xypolopoulos, C., & Vazirgiannis, M. (2018). Word Embeddings from Large-Scale Greek Web content. arXiv Preprint arXiv:1810.06694.
- Palomar-Giner, J., Saiz, J. J., Espuña, F., Mina, M., Da Dalt, S., Llop, J., Ostendorff, M., Ortiz Suarez, P., Rehm, G., Gonzalez-Agirre, A., & Villegas, M. (2024). A CURATEd CATalog: Rethinking the Extraction of Pretraining Corpora for Mid-Resourced Languages. In N. Calzolari, M.-Y. Kan, V. Hoste, A. Lenci, S. Sakti, & N. Xue (Eds.), Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024) (pp. 335–349). ELRA and ICCL. [Link](https://aclanthology.org/2024.lrec-main.31)
- Papaloukas, C., Chalkidis, I., Athinaios, K., Pantazi, D.-A., & Koubarakis, M. (2021). Multi-granular Legal Topic Classification on Greek Legislation. Proceedings of the Natural Legal Language Processing Workshop 2021, 63–75. [Link](https://doi.org/10.48550/arXiv.2109.15298)
- Popa-Fabre, M., Ortiz Suárez, P. J., Sagot, B., & de la Clergerie, É. (2020). French Contextualized Word-Embeddings with a sip of CaBeRnet: A New French Balanced Reference Corpus. Proceedings of the 8th Workshop on Challenges in the Management of Large Corpora, 15–23. [Link](https://aclanthology.org/2020.cmlc-1.3)
- Rae, J. W., Potapenko, A., Jayakumar, S. M., Hillier, C., & Lillicrap, T. P. (2019). Compressive Transformers for Long-Range Sequence Modelling. arXiv Preprint. [Link](https://arxiv.org/abs/1911.05507)
- Rodrigues, J., Gomes, L., Silva, J., Branco, A., Santos, R., Cardoso, H. L., & Osório, T. (2023). Advancing Neural Encoding of Portuguese with Transformer Albertina PT-\*.
- Rødven-Eide, S. (2016). The Swedish Culturomics Gigaword CorpusThe Swedish Culturomics Gigaword Corpus [Dataset]. Språkbanken Text. [Link](https://doi.org/10.23695/3WMV-1Z09)
- Sharma, E., Li, C., & Wang, L. (2019). BIGPATENT: A Large-Scale Dataset for Abstractive and Coherent Summarization. CoRR, abs/1906.03741. [Link](http://arxiv.org/abs/1906.03741)
- Soldaini, L., & Lo, K. (2023). peS2o (Pretraining Efficiently on S2ORC) Dataset. Allen Institute for AI.
- Strømberg-Derczynski, L., Ciosici, M., Baglini, R., Christiansen, M. H., Dalsgaard, J. A., Fusaroli, R., Henrichsen, P. J., Hvingelby, R., Kirkedal, A., Kjeldsen, A. S., Ladefoged, C., Nielsen, F. Å., Madsen, J., Petersen, M. L., Rystrøm, J. H., & Varab, D. (2021). The Danish Gigaword Corpus. Proceedings of the 23rd Nordic Conference on Computational Linguistics (NoDaLiDa), 413–421. [Link](https://aclanthology.org/2021.nodalida-main.46)
- Subramani, N., Luccioni, S., Dodge, J., & Mitchell, M. (2023). Detecting Personal Information in Training Corpora: An Analysis. 208–220. [Link](https://doi.org/10.18653/v1/2023.trustnlp-1.18)
- Varab, D., & Schluter, N. (2020). DaNewsroom: A Large-scale Danish Summarisation Dataset. Proceedings of The 12th Language Resources and Evaluation Conference, 6731–6739. [Link](https://www.aclweb.org/anthology/2020.lrec-1.831)
- Váradi, T., Nyéki, B., Koeva, S., Tadić, M., Štefanec, V., Ogrodniczuk, M., Nitoń, B., Pezik, P., Barbu Mititelu, V., Irimia, E., Mitrofan, M., Tufi\textcommabelows, D., Garabík, R., Krek, S., & Repar, A. (2022). Introducing the CURLICAT Corpora: Seven-language Domain Specific Annotated Corpora from Curated Sources. In N. Calzolari, F. Béchet, P. Blache, K. Choukri, C. Cieri, T. Declerck, S. Goggi, H. Isahara, B. Maegaard, J. Mariani, H. Mazo, J. Odijk, & S. Piperidis (Eds.), Proceedings of the Thirteenth Language Resources and Evaluation Conference (pp. 100–108). European Language Resources Association. [Link](https://aclanthology.org/2022.lrec-1.11)
- Wagner Filho, J. A., Wilkens, R., Idiart, M., & Villavicencio, A. (2018). The brwac corpus: A new open resource for brazilian portuguese. Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018).
- Žagar, A., Kavaš, M., Robnik-Šikonja, M., Erjavec, T., Fišer, D., Ljubešić, N., Ferme, M., Borovič, M., Boškovič, B., Ojsteršek, M., & Hrovat, G. (2022). Corpus of academic Slovene KAS 2.0. [Link](http://hdl.handle.net/11356/1448)
- Alicia Parrish, Angelica Chen, Nikita Nangia, Vishakh Padmakumar, Jason Phang, Jana Thompson, Phu Mon Htut, and Samuel Bowman. 2022. BBQ: A hand-built bias benchmark for question answering. In Findings of the Association for Computational Linguistics: ACL 2022, pages 2086–2105, Dublin, Ireland. Association for Computational Linguistics.
- Emily Sheng, Kai-Wei Chang, Premkumar Natarajan, and Nanyun Peng. 2019. The Woman Worked as a Babysitter: On Biases in Language Generation. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), pages 3407–3412, Hong Kong, China. Association for Computational Linguistics.
- Clark, P., Cowhey, I., Etzioni, O., Khot, T., Sabharwal, A., Schoenick, C., & Tafjord, O. (2018). Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge. arXiv:1803. 05457v1.
- Richard Socher, Alex Perelygin, Jean Wu, Jason Chuang, Christopher D. Manning, Andrew Ng, and Christopher Potts. 2013. Recursive Deep Models for Semantic Compositionality Over a Sentiment Treebank. In Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing, pages 1631–1642, Seattle, Washington, USA. Association for Computational Linguistics.
</details>
</details>
The model was trained for 3 epochs, with two final rounds of 0.3B higher-quality tokens each,
meaning that the total number of tokens seen during pre-training amounts to roughly 7.8 trillion tokens.
We provide an extense Datasheet section following the best practices defined by [(Gebru et al., 2021)](https://arxiv.org/pdf/1803.09010).
<details>
<summary>Datasheet</summary>
#### Motivation
**For what purpose was the dataset created? Was there a specific task in mind? Was there a specific gap that needed to be filled? Please provide a description.**
The purpose of creating this dataset is to pre-train the Salamandra family of multilingual models with high performance in a large number of
European languages (35) and code (including 92 different programming languages). In addition, we aim to represent especially the co-official
languages of Spain: Spanish, Catalan, Galician, and Basque. This is the reason why we carry out an oversampling of these languages.
We detected that there is a great lack of massive multilingual data, especially in minority languages (Ostendorff & Rehm, 2023), so part of
our efforts in the creation of this pre-training dataset have resulted in the contribution to large projects such as the Community OSCAR
(Brack et al., 2024), which includes 151 languages and 40T words, or CATalog (Palomar-Giner et al., 2024), the largest open dataset in
Catalan in the world.
**Who created the dataset (e.g., which team, research group) and on behalf of which entity (e.g., company, institution, organization)?**
The dataset has been created by the Language Technologies unit (LangTech) of the Barcelona Supercomputing Center - Centro Nacional de
Supercomputación (BSC-CNS), which aims to advance the field of natural language processing through cutting-edge research and development
and the use of HPC. In particular, it was created by the unit's data team, the main contributors being Javier Saiz, Ferran Espuña, and
Jorge Palomar.
However, the creation of the dataset would not have been possible without the collaboration of a large number of collaborators, partners,
and public institutions, which can be found in detail in the acknowledgements.
**Who funded the creation of the dataset? If there is an associated grant, please provide the name of the grantor and the grant name and number.**
This work/research has been promoted and financed by the Government of Catalonia through the [Aina project](https://projecteaina.cat/).
#### Composition
**What do the instances that comprise the dataset represent (e.g., documents, photos, people, countries)? Are there multiple types of instances (e.g., movies, users, and ratings; people and interactions between them; nodes and edges)? Please provide a description.**
The dataset consists entirely of text documents in various languages. Specifically, data was mainly sourced from the following databases and
repositories:
- **Common Crawl:** Repository that holds website data and is run by the Common Crawl non-profit organization. It is updated monthly and is
distributed under the CC0 1.0 public domain license.
- **GitHub:** Community platform that allows developers to create, store, manage, and share their code. Repositories are crawled and then
distributed with their original licenses, which may vary from permissive to non-commercial licenses.
- **Wikimedia:** Database that holds the collection databases managed by the Wikimedia Foundation, including Wikipedia, Wikibooks, Wikinews,
Wikiquote, Wikisource, and Wikivoyage. It is updated monthly and is distributed under Creative Commons Attribution-ShareAlike License 4.0.
- **EurLex:** Repository that holds the collection of legal documents from the European Union, available in all of the EU’s 24 official
languages and run by the Publications Office of the European Union. It is updated daily and is distributed under the Creative Commons
Attribution 4.0 International license.
- **Other repositories:** Specific repositories were crawled under permission for domain-specific corpora, which include academic, legal,
and newspaper repositories.
We provide a complete list of dataset sources at the end of this section.
**How many instances are there in total (of each type, if appropriate)?**
The dataset contains a diverse range of instances across multiple languages, with notable adjustments for certain languages. English
represents the largest portion, accounting for 39.08% of the total data. Spanish was upsampled by a factor of 2, bringing its share to 16.59%,
while Catalan (1.84%), Basque (0.26%), and Galician (0.36%) were also upsampled by 2. On the other hand, code-related data was downsampled
by half, making up 6.42% of the total. Other prominent languages include French (6.59%), Russian (5.39%), German (4.25%), and Hungarian
(3.93%), with several additional languages contributing between 1% and 2%, and smaller portions represented by a variety of others.
**Does the dataset contain all possible instances or is it a sample (not necessarily random) of instances from a larger set? If the dataset is a sample, then what is the larger set? Is the sample representative of the larger set (e.g., geographic coverage)? If so, please describe how this representativeness was validated/verified. If it is not representative of the larger set, please describe why not (e.g., to cover a more diverse range of instances, because instances were withheld or unavailable).**
The dataset is a sample from multiple sources, with different weights based on the primary language of the content: Spanish, Catalan,
Basque, and Galician content was upsampled by a factor of two, while programming languages were downsampled by a factor of half. Other
sources were sampled in proportion to their occurrence.
**What data does each instance consist of? “Raw” data (e.g., unprocessed text or images) or features? In either case, please provide a description.**
Each instance consists of a text document processed for deduplication, language identification, and source-specific filtering. Some
documents required optical character recognition (OCR) to extract text from non-text formats such as PDFs.
**Is there a label or target associated with each instance? If so, please provide a description.**
Each instance is labeled with a unique identifier, the primary language of the content, and the URL for web-sourced instances. Additional
labels were automatically assigned to detect specific types of content —harmful or toxic content— and to assign preliminary indicators of
undesired qualities —very short documents, high density of symbols, etc.— which were used for filtering instances.
**Is any information missing from individual instances? If so, please provide a description, explaining why this information is missing (e.g., because it was unavailable). This does not include intentionally removed information, but might include, e.g., redacted text.**
No significant information is missing from the instances.
**Are relationships between individual instances made explicit (e.g., users’ movie ratings, social network links)? If so, please describe how these relationships are made explicit.**
Instances are related through shared metadata, such as source and language identifiers.
**Are there recommended data splits (e.g., training, development/validation, testing)? If so, please provide a description of these splits, explaining the rationale behind them.**
The dataset is split randomly into training, validation, and test sets.
**Are there any errors, sources of noise, or redundancies in the dataset? If so, please provide a description.**
Despite removing duplicated instances within each source, redundancy remains at the paragraph and sentence levels, particularly in
web-sourced instances where SEO techniques and templates contribute to repeated textual patterns. Some instances may also be duplicated
across sources due to format variations.
**Is the dataset self-contained, or does it link to or otherwise rely on external resources (e.g., websites, tweets, other datasets)? If it links to or relies on external resources, a) are there guarantees that they will exist, and remain constant, over time; b) are there official archival versions of the complete dataset (i.e., including the external resources as they existed at the time the dataset was created); c) are there any restrictions (e.g., licenses, fees) associated with any of the external resources that might apply to a dataset consumer? Please provide descriptions of all external resources and any restrictions associated with them, as well as links or other access points, as appropriate.**
The dataset is self-contained and does not rely on external resources.
**Does the dataset contain data that might be considered confidential (e.g., data that is protected by legal privilege or by doctor–patient confidentiality, data that includes the content of individuals’ non-public communications)? If so, please provide a description.**
The dataset does not contain confidential data.
**Does the dataset contain data that, if viewed directly, might be offensive, insulting, threatening, or might otherwise cause anxiety? If so, please describe why. If the dataset does not relate to people, you may skip the remaining questions in this section.**
The dataset includes web-crawled content, which may overrepresent pornographic material across languages (Kreutzer et al., 2022). Although
pre-processing techniques were applied to mitigate offensive content, the heterogeneity and scale of web-sourced data make exhaustive
filtering challenging, which makes it next to impossible to identify all adult content without falling into excessive filtering, which may
negatively influence certain demographic groups (Dodge et al., 2021).
**Does the dataset identify any subpopulations (e.g., by age, gender)? If so, please describe how these subpopulations are identified and provide a description of their respective distributions within the dataset.**
The dataset does not explicitly identify any subpopulations.
**Is it possible to identify individuals (i.e., one or more natural persons), either directly or indirectly (i.e., in combination with other data) from the dataset? If so, please describe how.**
Web-sourced instances in the dataset may contain personally identifiable information (PII) that is publicly available on the Web, such as
names, IP addresses, email addresses, and phone numbers. While it would be possible to indirectly identify individuals through the
combination of multiple data points, the nature and scale of web data makes it difficult to parse such information. In any case, efforts are
made to filter or anonymize sensitive data during pre-processing, but some identifiable information may remain in the dataset.
**Does the dataset contain data that might be considered sensitive in any way? If so, please provide a description.**
Given that the dataset includes web-sourced content and other publicly available documents, instances may inadvertently reveal financial
information, health-related details, or forms of government identification, such as social security numbers (Subramani et al., 2023),
especially if the content originates from less-regulated sources or user-generated platforms.
#### Collection Process
**How was the data collected?**
This dataset is constituted by combining several sources, whose acquisition methods can be classified into three groups:
- Web-sourced datasets with some preprocessing available under permissive license (p.e. Common Crawl).
- Domain-specific or language-specific raw crawls (p.e. Spanish Crawling).
- Manually curated data obtained through collaborators, data providers (by means of legal assignment agreements) or open source projects
(p.e. CATalog).
**What mechanisms or procedures were used to collect the data? How were these mechanisms or procedures validated?**
According to the three groups previously defined, these are the mechanisms used in each of them:
- Open direct download. Validation: data integrity tests.
- Ad-hoc scrapers or crawlers. Validation: software unit and data integrity tests.
- Direct download via FTP, SFTP, API or S3. Validation: data integrity tests.
**If the dataset is a sample from a larger set, what was the sampling strategy?**
The sampling strategy was to use the whole dataset resulting from the filtering explained in the ‘preprocessing/cleaning/labelling’ section,
with the particularity that an upsampling of 2 (i.e. twice the probability of sampling a document) was performed for the co-official
languages of Spain (Spanish, Catalan, Galician, Basque), and a downsampling of 1/2 was applied for code (half the probability of sampling a
code document, evenly distributed among all programming languages).
**Who was involved in the data collection process and how were they compensated?**
This data is generally extracted, filtered and sampled by automated processes. The code required to run these processes has been developed
entirely by members of the LangTech data team, or otherwise obtained from open-source software. Furthermore, there has been no monetary
consideration for acquiring data from suppliers.
**Over what timeframe was the data collected? Does this timeframe match the creation timeframe of the data associated with the instances? If not, please describe the timeframe in which the data associated with the instances was created.**
Data were acquired and processed from April 2023 to April 2024. However, as mentioned, much data has been obtained from open projects such
as Common Crawl, which contains data from 2014, so it is the end date (04/2024) rather than the start date that is important.
**Were any ethical review processes conducted? If so, please provide a description of these review processes, including the outcomes, as well as a link or other access point to any supporting documentation.**
No particular ethical review process has been carried out as the data is mostly open and not particularly sensitive. However, we have an
internal evaluation team and a bias team to monitor ethical issues. In addition, we work closely with ‘Observatori d'Ètica en Intel·ligència
Artificial’ (OEIAC) and ‘Agencia Española de Supervisión de la Inteligencia Artificial’ (AESIA) to audit the processes we carry out from an
ethical and legal point of view, respectively.
#### Preprocessing
**Was any preprocessing/cleaning/labeling of the data done? If so, please provide a description. If not, you may skip the remaining questions in this section.**
Instances of text documents were not altered, but web-sourced documents were filtered based on specific criteria along two dimensions:
- Quality: documents with a score lower than 0.8, based on undesired qualities, such as documents with low number of lines, very short
sentences, presence of long footers and headers, and high percentage of punctuation, obtained through CURATE (Palomar-Giner et al., 2024)
were filtered out.
- Harmful or adult content: documents originating from Colossal OSCAR were filtered using LLM-Datasets (Ostendorff et al., 2024) based on
the perplexity from a language model (‘harmful_pp’ field) provided by the Ungoliant pipeline (Abadji et al., 2021).
**Was the “raw” data saved in addition to the preprocessed/cleaned/labeled data? If so, please provide a link or other access point to the “raw” data.**
The original raw data was not kept.
**Is the software that was used to preprocess/clean/label the data available? If so, please provide a link or other access point.**
Yes, the preprocessing and filtering software is open-sourced. The [CURATE](https://github.com/langtech-bsc/CURATE) pipeline was used for Spanish Crawling and CATalog,
and the [Ungoliant](https://github.com/oscar-project/ungoliant) pipeline was used for the OSCAR project.
#### Uses
**Has the dataset been used for any tasks already? If so, please provide a description.**
Pre-train the Salamandra model family.
**What (other) tasks could the dataset be used for?**
The data can be used primarily to pre-train other language models, which can then be used for a wide range of use cases. The dataset could
also be used for other tasks such as fine-tuning language models, cross-lingual NLP tasks, machine translation, domain-specific text
generation, and language-specific data analysis.
**Is there anything about the composition of the dataset or the way it was collected and preprocessed/cleaned/labeled that might impact future uses? Is there anything a dataset consumer could do to mitigate these risks or harms?**
Web-crawled content is over-represented with standard language varieties, impacting language model performance for minority languages.
Language diversity in data is crucial to avoid bias, especially in encoding non-standard dialects, preventing the exclusion of demographic
groups. Moreover, despite legal uncertainties in web-scraped data, we prioritize permissive licenses and privacy protection measures,
acknowledging the challenges posed by personally identifiable information (PII) within large-scale datasets. Our ongoing efforts aim to
address privacy concerns and contribute to a more inclusive linguistic dataset.
**Are there tasks for which the dataset should not be used?**
-
#### Distribution
**Will the dataset be distributed to third parties outside of the entity on behalf of which the dataset was created? If so, please provide a description.**
The dataset will not be released or distributed to third parties. Any related question to distribution is omitted in this section.
#### Maintenance
**Who will be supporting/hosting/maintaining the dataset?**
The dataset will be hosted by the Language Technologies unit (LangTech) of the Barcelona Supercomputing Center (BSC). The team will ensure
regular updates and monitor the dataset for any issues related to content integrity, legal compliance, and bias for the sources they are
responsible for.
**How can the owner/curator/manager of the dataset be contacted?**
The data owner may be contacted with the email address langtech@bsc.es.
**Will the dataset be updated?**
The dataset will not be updated.
**If the dataset relates to people, are there applicable limits on the retention of the data associated with the instances? If so, please describe these limits and explain how they will be enforced.**
The dataset does not keep sensitive data that could allow direct identification of individuals, apart from the data that is publicly
available in web-sourced content. Due to the sheer volume and diversity of web data, it is not feasible to notify individuals or manage data
retention on an individual basis. However, efforts are made to mitigate the risks associated with sensitive information through
pre-processing and filtering to remove identifiable or harmful content. Despite these measures, vigilance is maintained to address potential
privacy and ethical issues.
**Will older versions of the dataset continue to be supported/hosted/maintained? If so, please describe how. If not, please describe how its obsolescence will be communicated to dataset consumers.**
Since the dataset will not be updated, only the final version will be kept.
**If others want to extend/augment/build on/contribute to the dataset, is there a mechanism for them to do so?**
The dataset does not allow for external contributions.
</details>
### Finetuning Data
This instruction-tuned variant has been trained with a mixture of 276k English, Spanish, and Catalan multi-turn instructions gathered from open datasets:
| Dataset | ca | en | es |
|-----------------------|:------:|:------:|:------:|
| alpaca-cleaned | - | 50,000 | - |
| aya-dataset | - | 3,944 | 3,854 |
| CoQCat | 4,797 | - | - |
| databricks-dolly-15k | - | 15,011 | - |
| dolly-3k-ca | 3,232 | - | - |
| flores-instr | 1,994 | 1,994 | 3,988 |
| MentorCA | 7,122 | - | - |
| MentorES | - | - | 7,122 |
| no-robots | - | 9,499 | - |
| oasst-ca | 2,518 | - | - |
| oasst2 | 750 | 31,086 | 15,438 |
| open-orca | - | 50,000 | - |
| RagMultilingual | 16,043 | 14,997 | 11,263 |
| tower-blocks | - | 19,895 | 2,000 |
| **Total** | **36,456** | **196,426** | **43,665** |
---
## Evaluation
### Gold-standard benchmarks
Evaluation is done using the Language Model Evaluation Harness (Gao et al., 2024). We evaluate on a set of tasks taken from [SpanishBench](https://github.com/EleutherAI/lm-evaluation-harness/pull/2157), [CatalanBench](https://github.com/EleutherAI/lm-evaluation-harness/pull/2154), [BasqueBench](https://github.com/EleutherAI/lm-evaluation-harness/pull/2153) and [GalicianBench](https://github.com/EleutherAI/lm-evaluation-harness/pull/2155). These benchmarks include both new and existing tasks and datasets. Given that this is an instructed model, we add LM Evaluation Harness's native feature of `chat-template` to the setup. In the tables below, we include the results in a selection of evaluation datasets that represent model's performance across a variety of tasks within these benchmarks.
We only use tasks that are either human generated, human translated, or with a strong human-in-the-loop (i.e., machine translation followed by professional revision or machine generation followed by human revision and annotation). This is the reason behind the variety in number of tasks reported across languages. As more tasks that fulfill these requirements are published, we will update the presented results. We also intend to expand the evaluation to other languages, as long as the datasets meet our quality standards.
During the implementation of the evaluation we observed a series of issues worth considering when replicating and interpreting the results presented. These issues include ≈1.5% variances in performance in some tasks depending on the version of the `transformers` library used, and depending on the use (or lack of use) of tensor parallelism when loading a model. When implementing existing tasks, we carry out a comprehensive quality evaluation of the dataset, the Harness task itself, and what kind of input models see during evaluation. Our implementation (see links above) addresses multiple existing problems such as errors in datasets and prompts, and lack of pre-processing. All this means that results will vary if using other Harness implementations, and may slightly vary depending on the replication setup.
It should be noted that these results are subject to all the drawbacks of every current gold-standard evaluation, and that the figures do not fully represent the models capabilities and potential. We thus advise caution when reading and interpreting the results.
A full list of results compared to other baselines, a discussion of the model's performance across tasks and its implications, and details regarding problem-solving with task implementation will soon be available in the technical report.
All results reported below are on a 0-shot setting.
#### Spanish
<table><thead>
<tr>
<th>Category</th>
<th>Task</th>
<th>Metric</th>
<th>Result</th>
</tr></thead>
<tbody>
<tr>
<td>Commonsense Reasoning</td>
<td>xstorycloze_es</td>
<td>acc</td>
<td>69.29</td>
</tr>
<tr>
<td rowspan="2">NLI</td>
<td>wnli_es</td>
<td>acc</td>
<td>45.07</td>
</tr>
<tr>
<td>xnli_es</td>
<td>acc</td>
<td>51.49</td>
</tr>
<tr>
<td>Paraphrasing</td>
<td>paws_es</td>
<td>acc</td>
<td>59.4</td>
</tr>
<tr>
<td>QA</td>
<td>xquad_es</td>
<td>acc</td>
<td>43.82</td>
</tr>
<tr>
<td>Translation</td>
<td>flores_es</td>
<td>bleu</td>
<td>22.98</td>
</tr>
</tbody>
</table>
#### Catalan
<table><thead>
<tr>
<th>Category</th>
<th>Task</th>
<th>Metric</th>
<th>Result</th>
</tr></thead>
<tbody>
<tr>
<td rowspan="2">Commonsense Reasoning</td>
<td>copa_ca</td>
<td>acc</td>
<td>81.2</td>
</tr>
<tr>
<td>xstorycloze_ca</td>
<td>acc</td>
<td>70.68</td>
</tr>
<tr>
<td rowspan="2">NLI</td>
<td>wnli_ca</td>
<td>acc</td>
<td>50.7</td>
</tr>
<tr>
<td>xnli_ca</td>
<td>acc</td>
<td>55.14</td>
</tr>
<tr>
<td rowspan="2">Paraphrasing</td>
<td>parafraseja</td>
<td>acc</td>
<td>65.18</td>
</tr>
<tr>
<td>paws_ca</td>
<td>acc</td>
<td>62.95</td>
</tr>
<tr>
<td rowspan="5">QA</td>
<td>arc_ca_easy</td>
<td>acc</td>
<td>64.98</td>
</tr>
<tr>
<td>arc_ca_challenge</td>
<td>acc</td>
<td>41.89</td>
</tr>
<tr>
<td>openbookqa_ca</td>
<td>acc</td>
<td>35.2</td>
</tr>
<tr>
<td>piqa_ca</td>
<td>acc</td>
<td>69.53</td>
</tr>
<tr>
<td>siqa_ca</td>
<td>acc</td>
<td>48.62</td>
</tr>
<tr>
<td>Translation</td>
<td>flores_ca</td>
<td>bleu</td>
<td>28.65</td>
</tr>
</tbody></table>
#### Basque
<table><thead>
<tr>
<th>Category</th>
<th>Task</th>
<th>Metric</th>
<th>Result</th>
</tr></thead>
<tbody>
<tr>
<td rowspan="2">Commonsense Reasoning</td>
<td>xcopa_eu</td>
<td>acc</td>
<td>61.6</td>
</tr>
<tr>
<td>xstorycloze_eu</td>
<td>acc</td>
<td>61.15</td>
</tr>
<tr>
<td rowspan="2">NLI</td>
<td>wnli_eu</td>
<td>acc</td>
<td>45.07</td>
</tr>
<tr>
<td>xnli_eu</td>
<td>acc</td>
<td>46.81</td>
</tr>
<tr>
<td rowspan="3">QA</td>
<td>eus_exams</td>
<td>acc</td>
<td>39.09</td>
</tr>
<tr>
<td>eus_proficiency</td>
<td>acc</td>
<td>36.93</td>
</tr>
<tr>
<td>eus_trivia</td>
<td>acc</td>
<td>46.94</td>
</tr>
<tr>
<td>Reading Comprehension</td>
<td>eus_reading</td>
<td>acc</td>
<td>45.45</td>
</tr>
<tr>
<td>Translation</td>
<td>flores_eu</td>
<td>bleu</td>
<td>14.89</td>
</tr>
</tbody></table>
#### Galician
<table><thead>
<tr>
<th>Category</th>
<th>Task</th>
<th>Metric</th>
<th>Result</th>
</tr></thead>
<tbody>
<tr>
<td rowspan="2">Paraphrasing</td>
<td>parafrases_gl</td>
<td>acc</td>
<td>55.44</td>
</tr>
<tr>
<td>paws_gl</td>
<td>acc</td>
<td>56.55</td>
</tr>
<tr>
<td>QA</td>
<td>openbookqa_gl</td>
<td>acc</td>
<td>38.4</td>
</tr>
<tr>
<td>Translation</td>
<td>flores_gl</td>
<td>bleu</td>
<td>27.03</td>
</tr>
</tbody>
</table>
### LLM-as-a-judge
We use [Prometheus-2 8x7B](https://huggingface.co/prometheus-eval/prometheus-8x7b-v2.0) as a judge to evaluate the responses of the model. Tasks are created from existing multilingual evaluation datasets covering the same categories as the ones measured in our gold-standard benchmarks. We randomly select a subset of 250 instances per language from the `test` set of each source dataset. To evaluate the responses of our model, we use task-specific criteria developed in-house for the _LLM-judge_ to use. Each criterion is measured either as a 5-point Likert scale or as a binary task depending on the idiosyncrasy of the task and criterion.
Prompts for each task are created in various ways to score the model's robustness in addition to these criteria. This is done by presenting the same source instance within three different prompts. We then calculate the variance between the scores assigned by the _LLM-judge_ to our model's responses to the three prompt styles and average it across all instances. Prompts are human translated to all languages measured. We do not provide the _LLM-judge_ with a reference answer.
The _judge_ prompt we use during evaluation is the same used to fine tune the Prometheus-2 family. We keep the _judge_ prompt and criteria used to present the _LLM-judge_ with the task prompts and model responses in English for evaluation across languages. The _judge_ prompt used is:
```python
"You are a fair judge assistant tasked with providing clear, objective feedback based on specific criteria, ensuring each assessment reflects the absolute standards set for performance.
###Task Description:
An instruction (might include an Input inside it), a response to evaluate, and a score rubric representing a evaluation criteria are given.
1. Write a detailed feedback that assess the quality of the response strictly based on the given score rubric, not evaluating in general.
2. After writing a feedback, write a score that is an integer between {a} and {b}. You should refer to the score rubric.
3. The output format should look as follows: \"Feedback: (write a feedback for criteria) [RESULT] (an integer number between {a} and {b})\"
4. Please do not generate any other opening, closing, and explanations.
###The instruction to evaluate:
{input}
###Response to evaluate:
{prediction}
###Score Rubrics:
{criteria}
###Feedback:"
```
As an example, prompts for the Math task in English are based on instances from [MGSM](https://huggingface.co/datasets/juletxara/mgsm), and each instance is presented within these prompts:
```python
"en": [
("I need help with this math problem: \"", "\" Give me the answer step by step and also the final result separately."),
("Can you please help me answer this? \"", "\" Explain the answer and give me the final result as well. Thanks."),
("Help me with this problem: \"", "\" I need the answer explained and the final result separately.")
]
```
This task is then evaluated by the _LLM-judge_ using two criteria, reasoning capability (5-point Likert) and mathematical correctness (binary):
```python
reasoning_capability_criteria = {
"reasoning_capability": """
[Does the model's answer demonstrate reasoning capability?]
Score 1: The answer demonstrates poor reasoning, with illogical arguments or conclusions that do not follow from the provided information.
Score 2: The answer shows weak reasoning, with some logical connections but also contains significant flaws or gaps in the argumentation.
Score 3: The answer demonstrates adequate reasoning, with generally logical arguments, but may have minor flaws or a lack of depth in the reasoning process.
Score 4: The answer shows strong reasoning, with well-structured arguments and conclusions that logically follow from the information provided.
Score 5: The answer demonstrates exceptional reasoning, with clear, coherent, and insightful arguments that are logically sound and well-supported by the information provided."""
}
mathematical_correctness_binary_criteria = {
"mathematical_correctness_binary": """
[Is the model's answer mathematically correct?]
Score 0: The answer contains mathematical errors that render the solution incorrect or unreliable.
Score 1: The answer is mathematically correct, with accurate calculations and appropriate use of mathematical concepts."""
}
```
#### Multilingual results
Here, we present results for seven categories of tasks in Spanish, Catalan, Basque, Galician, and English. Results are presented for each task, criterion and language. Criteria with a `(B)` after their name are binary criteria (i.e., numbers go from 0 to 1, where 1 is best). The rest of the criteria are measured using a 5-point Likert scale, where 5 is best. The first number of the pair of numbers separated by `/` shows the average score for the criterion (and language). The second number of each pair is the robustness score, where numbers closer to 0 mean that the model generates similar responses when comparing the three prompt varieties for a single instance.
Further details on all tasks and criteria, a full list of results compared to other baselines, a discussion of the model's performance across tasks and its implications, and details regarding problem-solving with task implementation will soon be available in the technical report.
---
## Ethical Considerations and Limitations
We examine the presence of undesired societal and cognitive biases present in this model using different benchmarks. For societal biases,
we test performance using the BBQ dataset (Parrish et al., 2022) in the original English and the Regard dataset (Sheng et al., 2019).
We report that while performance is high (accuracies around 0.8 depending on the social category) in disambiguated settings,
the model performs very poorly in ambiguous settings, which indicates the presence of societal biases that need to be further addressed in post-training phases.
Our cognitive bias analysis focuses on positional effects in 0-shot settings, and majority class bias in few-shot settings.
For positional effects, we leverage the ARC Multiple Choice Question dataset (Clark et al., 2018). We observe significant,
but relatively weak primacy effects, whereby the model shows a preference for answers towards the beginning of the list of provided answers.
We measure effects of majority class effects in few-shot settings using SST-2 (Socher et al., 2013). We again detect significant effects,
with a small effect size. This suggests that the model is relatively robust against the examined cognitive biases.
We highlight that our analyses of these biases are by no means exhaustive and are limited by the relative scarcity of adequate resources
in all languages present in the training data. We aim to gradually extend and expand our analyses in future work.
These results can be expected from a model that has undergone only a preliminary instruction tuning.
These tests are performed in order to show the biases the model may contain. We urge developers to take
them into account and perform safety testing and tuning tailored to their specific applications of the model.
---
## Additional information
### Author
The Language Technologies Unit from Barcelona Supercomputing Center.
### Contact
For further information, please send an email to <langtech@bsc.es>.
### Copyright
Copyright(c) 2024 by Language Technologies Unit, Barcelona Supercomputing Center.
### Funding
This work has been promoted and financed by the Government of Catalonia through the [Aina Project](https://projecteaina.cat/).
This work is funded by the _Ministerio para la Transformación Digital y de la Función Pública_ - Funded by EU – NextGenerationEU
within the framework of [ILENIA Project](https://proyectoilenia.es/) with reference 2022/TL22/00215337, 2022/TL22/00215336, 2022/TL22/00215335, 2022/TL22/00215334.
### Acknowledgements
This project has benefited from the contributions of numerous teams and institutions, mainly through data contributions, knowledge transfer or technical support.
In Catalonia, many institutions have been involved in the project. Our thanks to Òmnium Cultural, Parlament de Catalunya, Institut d'Estudis Aranesos, Racó Català, Vilaweb, ACN, Nació Digital, El món and Aquí Berguedà.
At national level, we are especially grateful to our ILENIA project partners: CENID, HiTZ and CiTIUS for their participation. We also extend our genuine gratitude to the Spanish Senate and Congress, Fundación Dialnet, Fundación Elcano and the ‘Instituto Universitario de Sistemas Inteligentes y Aplicaciones Numéricas en Ingeniería (SIANI)’ of the University of Las Palmas de Gran Canaria.
At the international level, we thank the Welsh government, DFKI, Occiglot project, especially Malte Ostendorff, and The Common Crawl Foundation, especially Pedro Ortiz, for their collaboration. We would also like to give special thanks to the NVIDIA team, with whom we have met regularly, specially to: Ignacio Sarasua, Adam Henryk Grzywaczewski, Oleg Sudakov, Sergio Perez, Miguel Martinez, Felipes Soares and Meriem Bendris. Their constant support has been especially appreciated throughout the entire process.
Their valuable efforts have been instrumental in the development of this work.
### Disclaimer
Be aware that the model may contain biases or other unintended distortions.
When third parties deploy systems or provide services based on this model, or use the model themselves,
they bear the responsibility for mitigating any associated risks and ensuring compliance with applicable regulations,
including those governing the use of Artificial Intelligence.
The Barcelona Supercomputing Center, as the owner and creator of the model, shall not be held liable for any outcomes resulting from third-party use.
### Citation
Technical report and paper coming soon.
### License
[Apache License, Version 2.0](https://www.apache.org/licenses/LICENSE-2.0)
## Model Index
|Model|Base|Instruct|
|:---:|:---:|:---:|
|2B| [Link](https://huggingface.co/projecte-aina/salamandra-2b) | [Link](https://huggingface.co/projecte-aina/salamandra-2b-instruct) |
|7B| [Link](https://huggingface.co/projecte-aina/salamandra-7b) | [Link](https://huggingface.co/projecte-aina/salamandra-7b-instruct) |
|40B| WiP | WiP |
|