Spaces:

Viani
/

SimMart

Sleeping

App Files Files Community

SimMart / notebooks

271 kB

Ctrl+K

Ctrl+K

1 contributor

History: 2 commits

Viani's picture

Hackathon: GRPO single-GPU notebook with embedded curves + eval table (50 steps, GRPO step-040 beats SFT-init +10.6%)

5a4d0ef verified 12 days ago

hackathon_grpo_single_gpu.ipynb

271 kB
Hackathon: GRPO single-GPU notebook with embedded curves + eval table (50 steps, GRPO step-040 beats SFT-init +10.6%) 12 days ago