jayzou3773/less-is-moe-s1-calibration-128-seq8192 download history

jayzou3773/less-is-moe-s1-calibration-128-seq8192 is a text generation dataset on the Hugging Face Hub. In the last 30 days it was downloaded 35 times (9 in the last 7 days), and 193 times in total. It ranks #348,137 among datasets by monthly downloads.

Less-is-MoE S1K calibration data — 128 samples, seq_length 8192 This is the fixed calibration artifact used to prune GPT-OSS-120B, Qwen3.5-122B-A10B, and the Gemma-4-26B-A4B causal language tower. It uses the same 128 source rows as the full-length variant: yentinglin/s1K-1.1-trl-form

Models trained on less-is-moe-s1-calibration-128-seq8192

9 models list it as training data.

Open jayzou3773/less-is-moe-s1-calibration-128-seq8192 on Hugging Face

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