TsinghuaC3I/ZEDA download history

TsinghuaC3I/ZEDA is a text generation dataset on the Hugging Face Hub. In the last 30 days it was downloaded 135 times (28 in the last 7 days), and 813 times in total. It ranks #120,262 among datasets by monthly downloads.

ZEDA: Zero-Expert Self-Distillation Adaptation This repository contains the prompts and rollout data used in the paper "Post-Trained MoE Can Skip Half Experts via Self-Distillation". Zero-Expert Self-Distillation Adaptation (ZEDA) is a framework that transforms post-trained static Mix

Open TsinghuaC3I/ZEDA on Hugging Face

Sister project: Paper Pulse, the upvote history of every Hugging Face Daily Paper.