TranscenderNing

RuntimeError: CUDA error: invalid device ordinal

1

在运行器件报错：Traceback (most recent call last): File "/home/ldn/ChatGLM130Dir/GLM-130B/generate.py", line 218, in args = initialize(extra_args_provider=add_generation_specific_args) File "/home/ldn/ChatGLM130Dir/GLM-130B/initialize.py", line 51, in initialize **args = get_args(args_list)** File "/home/ldn/anaconda3/envs/glm1301/lib/python3.9/site-packages/SwissArmyTransformer/arguments.py", line 385, in get_args initialize_distributed(args) File...

Addvera

2

### PR types New features ### PR changes add a new finetune method in peft/vera ### Description 数据集webnlg + llama2 测试结果 

[P1] GPU Memory usage issue

2

 Why is the parameter count only 0.03%, yet the memory usage during training reaches over 60 GB, whereas Lora training usually requires only around 17 GB?

add reft method

5

### PR types New features ### PR changes Add reft in paddlenlp/peft/reft reft ├── pareft │ ├── config.py reft配置文件，继承pavenv.config │ ├── dataset.py reft数据处理，reft方法输入会有一个新的intervention_locations字段，表示干预的token的位置，例如f5+l5表示干预输入的前5个tokne和后5个token │ ├── init.py │ ├── interventions.py 干预网络...