KirillR/BearAtlas
BearAtlas Dataset BearAtlas is a synthetic russian-language dataset created to efficiently train LLMs on high-quality, structured, and semantically meaningful texts. It is based on the Cultura-Ru-Edu corpus, pre-filtered and further processed using modern clustering, information value assessment, and genre-based text generation methods. ๐ Quick Reference Source: Filtered data from the Cultura-Ru-Edu corpus Methods: Sentence-BERT embeddings โ UMAP โ DBSCAN +โฆ See the full description on the dataset page: https://huggingface.co/datasets/KirillR/BearAtlas.
| Author / organization | KirillR |
|---|---|
| License (as declared) | apache-2.0 |
| Downloads | 236 (as reported by Hugging Face) |
| Likes | 2 |
| Created | 20 May 2025 |
| Last updated at source | 30 June 2025 |
Popularity, last 2 days
236 now (-7 since 2026-10-16). Counts as reported by the source, snapshotted daily by Silytics.
Recent changes
- 16 Oct 2026 Description: BearAtlas Dataset BearAtlas is a synthetic russian-language dataset created to efficiently train LLMs on high-quality, structured, and semantically meaningful texts. It is based on the Cultura-Ru-Edu corpus, pre-filtered and further process โ BearAtlas Dataset BearAtlas is a synthetic russian-language dataset created to efficiently train LLMs on high-quality, structured, and semantically meaningful texts. It is based on the Cultura-Ru-Edu corpus, pre-filtered and further process
Changes detected by Silytics between refreshes of the source.
More from KirillR
Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.