qingy2024/SynGen-Antiloop-DPO download history
qingy2024/SynGen-Antiloop-DPO is a dataset on the Hugging Face Hub. In the last 30 days it was downloaded 105 times (12 in the last 7 days), and 473 times in total. It ranks #109,997 among datasets by monthly downloads.
10,000 rows of DPO data for preventing looping and ensuring high quality reasoning trace generation. Negative samples are a mix of: 30% synthetic generated looping. 70% lower quality reasoning from SynGen Nano. The chosen column data is all generated by DeepSeek R1 0528. Dataset format: I fo