search

QwenLM / Qwen2.5

Qwen2.5 is the large language model series developed by Qwen team, Alibaba Cloud.
8.22k stars 501 forks source link

在fine tune Qwen2-7B-Instruct 保存时候错误 #835

Open JHaoGao opened 1 month ago

JHaoGao commented 1 month ago

image 使用的命令是： bash finetune.sh -m models/models--Qwen--Qwen2-7B-Instruct/snapshots/41c66b0be1c3081f13defc6bdf946c2ef240d6a6 -d example_set.jsonl --deepspeed ds_config_zero2.json --use_lora True --q_lora True

其中ds_config_zero2.json, finetune.sh, finetune.py是在repo的example/sft里面找到的

使用的配置：

SBATCH --ntasks= 1

SBATCH --cpus-per-task= 1

SBATCH --mem=48G

SBATCH --gres=gpu:a100:4

JHaoGao commented 1 month ago

image It should be .yaml here? https://github.com/QwenLM/Qwen2/blob/main/examples/sft/llama-factory/finetune-zh.md

SuperAZHE commented 1 month ago

i think llamafactory-cli cmd should use yaml file，not json file。reference：https://github.com/hiyouga/LLaMA-Factory/blob/main/examples/README.md

github-actions[bot] commented 2 days ago

This issue has been automatically marked as inactive due to lack of recent activity. Should you believe it remains unresolved and warrants attention, kindly leave a comment on this thread.