You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
使用预训练能使用Lora的方式微调吗?还是说使用lora只能选择指令微调?我目前选择了lora微调,训练方式使用了预训练,模型用的本地的qwen2.5-7b-instruct,训练没有报错,但是训练完成之后loss是上升的,而不是下降,从效果上来看也没有生效。是发生了其它什么事情吗?倒反天罡了/捂脸
我的训练文件格式是这样的:
[
{"text": "# 所属代码域:channels/dynamic_barrier_value.py\ndef consume(self) -> bool:\n if self.seen == self.names:\n self.seen = set()\n self.names = None\n return True\n return False"},
{"text": "# 所属代码域:channels/context.py\nfrom langgraph.managed.context import Context as ContextManagedValue\n\nContext = ContextManagedValue.of\n\n__all__ = ["Context"]"}
]
Beta Was this translation helpful? Give feedback.
All reactions