ViperEkura eba99e1f5e feat(model): 添加QK归一化和门控注意力支持		2026-01-05 16:14:44 +08:00
..
__init__.py	refactor(trainer): 优化trainer 结构	2025-12-07 21:23:05 +08:00
model_config.py	feat(model): 添加QK归一化和门控注意力支持	2026-01-05 16:14:44 +08:00
param_config.py	refactor(trainer): 优化trainer 结构	2025-12-07 21:23:05 +08:00
schedule_config.py	refactor(trainer): 优化trainer 结构	2025-12-07 21:23:05 +08:00
train_config.py	refactor(paralell): 优化并行设备指定方法	2025-12-26 20:54:33 +08:00