CEfanmin comments

Results 15 comments of


                                            CEfanmin

How to disable "wandb" while running finetune.py

> You can remove that argument in `Trainer`, or change where you want to report your logs, e.g.` report_to=["tensorboard"]`. In this case you will have to make sure that it...

Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass,RuntimeError: element 0 of tensors does not require grad and does not have a grad_fn

Me too.

finetune time

> How to use multiple GPUs？I have try CUDA_VISIBLE_DEVICES, no effect do you success?now

Trying to fine tune starcoderbase model using finetuning.oy - multiple GPUs

> I am trying to fine tune bigcode/starcoderbase model on compute A100 with 8 GPUs 80Gb VRAM. My initial steps are to adjust parameters. I get some impression that it...

Add workspaces support to HTTP backend

返回的中文也是乱码

[Bug]: 返回中文会出现乱码

> Hi @CEfanmin, thanks for your feedback. Since LLMLingua uses token-level prompt compression, it can indeed cause garbled text in some languages. You can try using LLMLingua-2. Thanks！

qwen2.5-instruct不支持tool calls

> v0.15.2 v0.15.1版本支持Qwen2.5-72B-ins的工具调用，但不支持流式返回

官方代码部署两个人调用会报错RuntimeError: probability tensor contains either inf, nan or element < 0 #102

File "/home/fanmin/qianwen/code/qwenLLM.py", line 56, in _call generated_ids = self.model.generate(model_inputs.input_ids, ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/fanmin/qianwen/lib64/python3.11/site-packages/torch/utils/_contextlib.py", line 115, in decorate_context return func(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^ File "/home/fanmin/qianwen/lib64/python3.11/site-packages/transformers/generation/utils.py", line 1592, in generate return self.sample( ^^^^^^^^^^^^ File...

CEfanmin

如何训练？