8 lines
192 B
YAML
8 lines
192 B
YAML
- mode=train
|
|
- task=countdown2345
|
|
- algorithm=grpo
|
|
- algorithm.training.curriculum_schedule=gaussian
|
|
- model=qwen15
|
|
- algorithm.training.max_steps=1600
|
|
- algorithm.training.vllm_mode=colocate
|