Dataset · HuggingFaceTB
Synthetic SFT instruction examples
HuggingFaceTB/smoltalk
1M synthetic SFT examples used to train the SmolLM2-Instruct family.
What's inside
A synthetic dataset for supervised fine-tuning (SFT) of language models, containing 1M examples used in creating the SmolLM2-Instruct family. Collected to close the gap with proprietary instruction data; English language, text and tabular format.
What you can do with it
- SFT fine-tuning of instruction LLMs
- LoRA training for chat models
- Comparison with public SFT sets
- Languages
- English
- Size
- 1M–10M rows
- Downloads this month
- 55.2K
- Likes
- 469
Models trained on it
- dphn/Dolphin3.0-Llama3.1-8B-GGUF 61.4K
- bartowski/cognitivecomputations_Dolphin3.0-Mistral-24B-GGUFText generation 48.7K
- bartowski/Dolphin3.0-Llama3.1-8B-GGUFText generation 28.8K
- bartowski/Dolphin3.0-Llama3.2-3B-GGUFText generation 20.5K
- bartowski/cognitivecomputations_Dolphin3.0-R1-Mistral-24B-GGUFText generation 10.8K
- bartowski/Dolphin3.0-Qwen2.5-3b-GGUFText generation 8.7K
- bartowski/Dolphin3.0-Qwen2.5-1.5B-GGUFText generation 7.8K
- bartowski/Dolphin3.0-Llama3.2-1B-GGUFText generation 6.1K
How to load it
from datasets import load_dataset
ds = load_dataset("HuggingFaceTB/smoltalk")