nampdn-ai/tiny-textbooks
Textbook-like Dataset: A High-Quality Resource for Small Language Models The idea is simply inspired by the Textbooks Are All You Need II: phi-1.5 technical report paper. The source texts in this dataset have been gathered and carefully select the best of the falcon-refinedweb and minipile datasets to ensure the diversity, quality while tiny in size. The dataset was synthesized using 4x3090 Ti cards over a period of 500 hours, thanks to Nous-Hermes-Llama2-13b finetuned model.… See the full description on the dataset page: https://huggingface.co/datasets/nampdn-ai/tiny-textbooks.
| Author / organization | nampdn-ai |
|---|---|
| License (as declared) | apache-2.0 |
| Downloads | 386 (as reported by Hugging Face) |
| Likes | 186 |
| Created | 10 August 2023 |
| Last updated at source | 3 July 2024 |
Popularity, last 2 days
386 now (-58 since 2026-10-16). Counts as reported by the source, snapshotted daily by Silytics.
Connected in the knowledge graph
- tensorblock/Nucleus-1B-alpha-1-GGUFModel · used to train
- NucleusOrg/Nucleus-1B-alpha-1Model · used to train
Recent changes
- 17 Oct 2026 Description: Textbook-like Dataset: A High-Quality Resource for Small Language Models The idea is simply inspired by the Textbooks Are All You Need II: phi-1.5 technical report paper. The source texts in this dataset have been gathered and carefully sel → Textbook-like Dataset: A High-Quality Resource for Small Language Models The idea is simply inspired by the Textbooks Are All You Need II: phi-1.5 technical report paper. The source texts in this dataset have been gathered and carefully sel
Changes detected by Silytics between refreshes of the source.
More from nampdn-ai
- nampdn-ai/tiny-codesDataset
- nampdn-ai/tiny-strange-textbooksDataset
- nampdn-ai/vietspeechDataset
- nampdn-ai/mini-zhDataset
- nampdn-ai/sample-ev-corpusDataset
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 Information may be incomplete or out of date.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.