datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
llama_factory__countdown_3arg__val_sample_evalsilicon-factory-sample
Silicon Factory Sample Dataset
Synthetic dataset generated by the Silicon Factory using local Qwen model.
Details
Entries: 50
Source: Synthetic (Qwen 2.5 0.5B)
License: MIT
Language: English
Usage
from datasets import load_dataset
dataset = load_dataset("AEUPH/silicon-factory-sample")
print(dataset["train"][0])
Generated With
Tree-Speculative Decoding
4D Brane Memory for consistency
Quality control & deduplication