Tobi Lütke 189916d6fb Move GRPO training out of default finetune pipeline vor 3 Monaten
..
accelerate_multi_gpu.yaml bf1b8fc90a lots of training stuff vor 3 Monaten
sft.yaml cbeeb1f89b Add wall-clock checkpoints and full eval defaults vor 3 Monaten
sft_local.yaml cbeeb1f89b Add wall-clock checkpoints and full eval defaults vor 3 Monaten