Jürgen Schmied

josch15366

·

AI & ML interests

None yet

Recent Activity

new activity about 2 hours ago

josch15366/Qwen3.8-Flash-Next-NVFP4-LocalHessian-Experts:Calibration corpus: what we got wrong, and the one thing worth copying (position stratification)

new activity about 2 hours ago

RadixArk/Qwen3.8-Flash-Next-NVFP4:Per-expert weight_scale_2 is worth ~0.9 % held-out NLL (and two contracts you got right)

updated a model about 9 hours ago

josch15366/Qwen3.8-Flash-Next-NVFP4-LocalHessian-Experts

View all activity

Organizations

None yet

models 3

josch15366/Qwen3.8-Flash-Next-NVFP4-LocalHessian-Experts

Updated 44 minutes ago

josch15366/Qwen3.8-Flash-Next-FP8-lm_head

Updated about 20 hours ago

josch15366/Qwen3.8-27B-DFlash2-FP8

2B • Updated 1 day ago • 192 • 1

datasets 0

None public yet