--- task_categories: - feature-extraction pretty_name: HPLT2-embeddings size_categories: - n>1T language: - sq - bg - ca - cs - da - de - es - et - el - eu - fi - fr - gl - ga - hr - hu - hy - is - it - lv - lt - mk - nl - pl - pt - ro - sl - sk - sr - tr - sv - nb - nn configs: - config_name: als_Latn data_files: - split: train path: als_Latn/* - config_name: bul_Cyrl data_files: - split: train path: bul_Cyrl/* - config_name: cat_Latn data_files: - split: train path: cat_Latn/* - config_name: ces_Latn data_files: - split: train path: ces_Latn/* - config_name: dan_Latn data_files: - split: train path: dan_Latn/* - config_name: deu_Latn data_files: - split: train path: deu_Latn/* - config_name: ekk_Latn data_files: - split: train path: ekk_Latn/* - config_name: ell_Grek data_files: - split: train path: ell_Grek/* - config_name: eus_Latn data_files: - split: train path: eus_Latn/* - config_name: fin_Latn data_files: - split: train path: fin_Latn/* - config_name: fra_Latn data_files: - split: train path: fra_Latn/* - config_name: gle_Latn data_files: - split: train path: gle_Latn/* - config_name: glg_Latn data_files: - split: train path: glg_Latn/* - config_name: hrv_Latn data_files: - split: train path: hrv_Latn/* - config_name: hun_Latn data_files: - split: train path: hun_Latn/* - config_name: hye_Armn data_files: - split: train path: hye_Armn/* - config_name: isl_Latn data_files: - split: train path: isl_Latn/* - config_name: ita_Latn data_files: - split: train path: ita_Latn/* - config_name: lit_Latn data_files: - split: train path: lit_Latn/* - config_name: lvs_Latn data_files: - split: train path: lvs_Latn/* - config_name: mkd_Cyrl data_files: - split: train path: mkd_Cyrl/* - config_name: nld_Latn data_files: - split: train path: nld_Latn/* - config_name: nno_Latn data_files: - split: train path: nno_Latn/* - config_name: nob_Latn data_files: - split: train path: nob_Latn/* - config_name: pol_Latn data_files: - split: train path: pol_Latn/* - config_name: por_Latn data_files: - split: train path: por_Latn/* - config_name: ron_Latn data_files: - split: train path: ron_Latn/* - config_name: slk_Latn data_files: - split: train path: slk_Latn/* - config_name: slv_Latn data_files: - split: train path: slv_Latn/* - config_name: spa_Latn data_files: - split: train path: spa_Latn/* - config_name: srp_Cyrl data_files: - split: train path: srp_Cyrl/* - config_name: swe_Latn data_files: - split: train path: swe_Latn/* - config_name: tur_Latn data_files: - split: train path: tur_Latn/* - config_name: ukr_Cyrl data_files: - split: train path: ukr_Cyrl/* --- # HPLT2-embeddings ## Dataset summary HPLT2-embeddings is an extension of the [**HPLT2**](https://hplt-project.org/datasets/v2.0) dataset, annotated with **document-level** [**Snowflake's Arctic-embed-m-v2.0**](https://huggingface.co/Snowflake/snowflake-arctic-embed-m-v2.0) **embeddings** for **35 languages**, making the dataset **useful for a variety of tasks**, including document clustering, filtering, and other multilingual research. Snowflake-arctic-embed-m-v2.0 has a sequence length limit of 8192 tokens, each document's embeddings are obtained by using the CLS token to embed each document. The embeddings were computed as part of our [**🦊 JQL: Judging Quality across Languages**](https://huggingface.co/spaces/JQL-AI/JQL) project and will be the basis for an upcoming high-quality subset of HPLT2. We believe that they can be useful for other multilingual research and applications. For more details, see our paper [Judging Quality Across Languages: A Multilingual Approach to Pretraining Data Filtering with Language Models](https://arxiv.org/abs/2505.22232). ## Usage You can load the dataset in Python using e.g.pandas: ```python import h5py import pandas as pd # Path to your .h5 file file_path = "000_001_00000.h5" # <-- Replace with your actual file path # Open the HDF5 file and load data with h5py.File(file_path, "r") as f: # Load the embeddings and document IDs from the "train" group embeddings = f["train/embeddings"][:] document_ids = f["train/document_id"][:] # Convert document IDs from bytes (if needed) if isinstance(document_ids[0], bytes): document_ids = [doc_id.decode("utf-8") for doc_id in document_ids] # Optionally: create a DataFrame (only if embeddings aren't too large for RAM) df = pd.DataFrame(embeddings) df.insert(0, "document_id", document_ids) # Add document_id as the first column # Preview the DataFrame print(df.head()) print(f"Loaded {len(df)} rows with shape {embeddings.shape[1]}-dimensional embeddings.") ``` ## Origin of the Dataset This dataset, derived from HPLT2, includes web content collected from 2013 to 2024. As HPLT2 is sourced from the broader internet, it may contain some personally identifiable information (PII), despite efforts to anonymize email addresses and public IP addresses during processing. ## Considerations for Data Usage For information on social impact, potential biases, and known limitations, please refer to the [HPLT2 documentation](https://hplt-project.org/datasets/v2.0). ## Citation information If you use this dataset in your research or applications, please use the following citation: ``` @article{ali2025judging, title = {Judging Quality Across Languages: A Multilingual Approach to Pretraining Data Filtering with Language Models}, author = { Mehdi Ali, Manuel Brack, Max Lübbering, Elias Wendt, Abbas Goher Khan, Richard Rutmann, Alex Jude, Maurice Kraus, Alexander Arno Weber, Felix Stollenwerk, David Kaczér, Florian Mai, Lucie Flek, Rafet Sifa, Nicolas Flores-Herr, Joachim Köhler, Patrick Schramowski, Michael Fromm, Kristian Kersting }, year = {2025}, journal = {arXiv preprint arXiv:2505:22232} } ```