Dataset · Hugging Face
ZeroAgency/ru-instruct-conversation-v1
Combined dataset of mostly Russian dialogs in form of conversations suitable for LLM fine-tuning scenarios. Total samples: 82208 Deduplicated using simhash(hamming_treshold=3). Datasets used: IlyaGusev/saiga_scored (min_score: 8, no bad by regexp) IlyaGusev/oasst2_ru_main_branch attn-signs/kolmogorov-3 attn-signs/russian-easy-instructions
| Author / organization | ZeroAgency |
|---|---|
| License (as declared) | mit |
| Downloads | 125 (as reported by Hugging Face) |
| Likes | 0 |
| Created | 18 March 2025 |
| Last updated at source | 18 March 2025 |
Popularity, last 2 days
125 now. Counts as reported by the source, snapshotted daily by Silytics.
More from ZeroAgency
Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 18 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.