InternVL2-26B模型微调后,报UnboundLocalError: local variable 'head_num' referenced before assignment错误
InternVL2-26B模型微调后,报UnboundLocalError: local variable 'head_num' referenced before assignment错误
我也在全量微调InternVL2-26B,没有出现上述问题。
same here. I also finetuned mini-internVL (Lora Finetuning). Facing the same error.. can someone help?
我也在全量微调InternVL2-26B,没有出现上述问题。
想请教一下,有没有InternVL2-26B全量微调的参考呀?
请教一下 有4张3090的卡 为啥这个26B模型只能在一张卡上跑
我也在全量微调InternVL2-26B,没有出现上述问题。
想请教一下,有没有InternVL2-26B全量微调的参考呀?
请参考:https://internvl.readthedocs.io/en/latest/internvl2.0/finetune.html#start-2nd-fine-tuning
请教一下 有4张3090的卡 为啥这个26B模型只能在一张卡上跑
请问您用的是什么框架
same here. I also finetuned mini-internVL (Lora Finetuning). Facing the same error.. can someone help?
Our new code has been released, please refer to https://internvl.readthedocs.io/en/latest/internvl2.0/finetune.html#start-2nd-fine-tuning.