Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

OOM during training #151

Open
Freddie1946 opened this issue Mar 11, 2025 · 2 comments
Open

OOM during training #151

Freddie1946 opened this issue Mar 11, 2025 · 2 comments

Comments

@Freddie1946
Copy link

Hello! I encountered an OOM issue during training. I am using 4×4090 (24GB). Does this fail to meet the minimum memory requirement?

Here are my training settings:

torchrun --nproc_per_node="4"
--nnodes="1"
--node_rank="0"
--master_addr="127.0.0.1"
--master_port="12346"
src/open_r1/grpo_rec.py
--deepspeed local_scripts/zero3.json
--output_dir output/$RUN_NAME
--model_name_or_path Qwen/Qwen2.5-VL-3B-Instruct
--dataset_name
--image_root
--max_prompt_length 1024
--num_generations 4
--per_device_train_batch_size 1
--gradient_accumulation_steps 2
--logging_steps 1
--bf16
--torch_dtype bfloat16
--data_seed 42
--report_to none
--gradient_checkpointing true
--attn_implementation flash_attention_2
--num_train_epochs 2
--run_name $RUN_NAME
--save_steps 100
--save_only_model true

@melodyangle
Copy link

melodyangle commented Mar 13, 2025

不够用的,需要6*24G。但GPU够用了,还是出现了内核问题,你的内核是多少;uname -r命令可以查看

@Freddie1946
Copy link
Author

不够用的,需要6*24G。但GPU够用了,还是出现了内核问题,你的内核是多少;uname -r命令可以查看

6.8.0-51-generic 是这个

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants