hotdogs/uka-fable-reasoning
Dataset em destaque no Hugging Face — 259 downloads. 🧠 uka-fable-reasoning v2.1 Curated reasoning dataset for fine-tuning LLMs as autonomous agents. 7,482 rows — ChatML format, multi-turn conversati…
O dataset hotdogs/uka-fable-reasoning está entre os destaques do Hugging Face — dados que alimentam o treinamento e a avaliação dos modelos do momento.
Ficha do dataset
- Idiomas: inglês
- Licença: AGPL-3.0
- Downloads: 259 · Curtidas: 11
Sobre o dataset
🧠 uka-fable-reasoning v2.1 Curated reasoning dataset for fine-tuning LLMs as autonomous agents. 7,482 rows — ChatML format, multi-turn conversations, all quality issues fixed. 📊 Dataset Overview Split Rows Messages Size (JSONL) Size (Parquet) Train 6,359 3–137 49.4 MB 22.1 MB Valid 748 3–10+ 5.8 MB 2.5 MB Test 375 3–10+ 3.0 MB 1.3 MB Total 7,482 58.2 MB 25.9 MB Conversation Types Type Count Description… See the full description on the dataset page: https://huggingface.co/datasets/hotdogs/uka-fable-reasoning.
Como carregar
Use a biblioteca datasets do Hugging Face:
pip install -U datasets
from datasets import load_dataset
ds = load_dataset("hotdogs/uka-fable-reasoning")
print(ds)
print(ds["train"][0])Tags
reasoning agentic sft chain-of-thought multi-turn tool-use chatml
Leia também
Linq’s iMessage Apps Bring Payments, Tickets, Flights, and Games Into the iMessage Bubble Through the imessage_app Part
The DeepMind trio who built a poker AI are now making money for quant hedge funds
