Yaowei Zheng

icon indicating an email [email protected]

icon company Millennium Science School icon location Beijing, China icon location No code All live

Results 651 comments of Yaowei Zheng

H20多卡训练batch大于1学习率一直为0

see https://github.com/hiyouga/LLaMA-Factory/issues/6201

Training Qwen/Qwen2.5-Coder-32B-Instruct model OOM

Have you tried `enable_liger_kernel: true`?

freeze微调报警告None of the inputs have requires_grad=True. Gradients will be None

训练命令？

CUDA initialization failed with distributed training on 8*B200 GPUs

Verify that the NCCL library and PyTorch are functioning.

llama3添加系统提示词后回答出现Human: Assistant:

see FAQs

ValueError: Failed to convert pandas DataFrame to Arrow Table from file

Make sure all your inputs are string type

when would you show how to fintune the Qwen2.5-VL model?

It is the same as the Qwen2-VL model: https://github.com/QwenLM/Qwen2.5-VL/tree/35ba6e18636510de4bf8d4a7caaca3f4f5163a84?tab=readme-ov-file#training

RuntimeError: Invalid buffer size: when try sft

Use transformers 4.49.0

RuntimeError: Invalid buffer size: when try sft

remove preprocessing num workers

Error during git lfs fetch: "Git LFS is disabled for this repository"

try `git lfs install`