Files
SkillOpt/configs/sealqa/default.yaml
T
2026-05-08 18:12:45 +00:00

24 lines
342 B
YAML

_base_: ../_base_/default.yaml
model:
reasoning_effort: medium
train:
batch_size: 10
accumulation: 1
gradient:
minibatch_size: 8
merge_batch_size: 8
optimizer:
learning_rate: 4
env:
name: sealqa
skill_init: reflact/envs/sealqa/skills/initial.md
split_dir: data/sealqa_split
workers: 4
max_tool_turns: 12
limit: 0