mlabonne/orca-agentinstruct-1M-v1-cleaned
๐ Orca-AgentInstruct-1M-v1-cleaned This is a cleaned version of the microsoft/orca-agentinstruct-1M-v1 dataset released by Microsoft. orca-agentinstruct-1M-v1 is a fully synthetic dataset using only raw text publicly available on the web as seed data. It is a subset of the full AgentInstruct dataset (~25M samples) that created Orca-3-Mistral. Compared to Mistral 7B Instruct, the authors claim 40% improvement on AGIEval, 19% improvement on MMLU, 54% improvement on GSM8K, 38%โฆ See the full description on the dataset page: https://huggingface.co/datasets/mlabonne/orca-agentinstruct-1M-v1-cleaned
| Author / organization | mlabonne |
|---|---|
| License (as declared) | cdla-permissive-2.0 |
| Downloads | 146 (as reported by Hugging Face) |
| Likes | 70 |
| Created | 17 November 2024 |
| Last updated at source | 25 January 2025 |
Popularity, last 2 days
146 now (-7 since 2026-10-16). Counts as reported by the source, snapshotted daily by Silytics.
Connected in the knowledge graph
- mradermacher/Llama-3.1-8B-Reason-Blend-888k-GGUFModel ยท used to train
- mradermacher/Llama-3.1-8B-Reason-Blend-888k-i1-GGUFModel ยท used to train
More from mlabonne
- mlabonne/harmless_alpacaDataset
- mlabonne/harmful_behaviorsDataset
- mlabonne/FineTome-100kDataset
- mlabonne/open-perfectblendDataset
- mlabonne/guanaco-llama2-1kDataset
- mlabonne/orpo-dpo-mix-40kDataset
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 Information may be incomplete or out of date.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.