-
python3.8 infer.py \
--base_model './llama-7b-hf' \
--lora_weights './lora-llama-med' \
--use_lora True \
--instruct_dir './data/infer.json' \
--prompt_template 'med_template'…
-
{'loss': 0.5511, 'learning_rate': 6.013071895424836e-06, 'epoch': 9.82}
{'loss': 0.5611, 'learning_rate': 4.967320261437908e-06, 'e…
-
请问一下，测试用的环境是什么啊。我是用的python3.10.11，安装完环境，cuda（版本为11.7）后运行infer.py
`python infer.py --base_model 'decapoda-research/llama-7b-hf' --lora_weights './lora-llama-med' --use_lora True --instruct_dir './d…
-
完整的报错信息如下：
nohup: ignoring input
===================================BUG REPORT===================================
Welcome to bitsandbytes. For bug reports, please submit your error trace to: http…
-
运行命令
accelerate launch --config_file /home/nvidia/.cache/huggingface/accelerate/deepspeed.yaml src/train_bash.py \
--stage sft \
--model_name_or_path baichuan-inc/Baichuan-13B-Chat \
-…
-
我很纳闷，为什么两个项目内容一模一样
https://github.com/focus-creative-games/hybridclr
https://github.com/tuyoogame/huatuo
-
Dear contributors,
Thanks for your insightful work and open-source project. I am happy to share our project (https://github.com/FreedomIntelligence/LLMZoo) with you, which includes a tailored vers…
-
请问哪里可以下载Huatuo-200K数据集？
-
-
I want to download the huatuoGPT model, but it seems that there is no，https://huggingface.co/FreedomIntelligence/HuatuoGPT-7b-v1