jianzhnie / LLamaTuner

Easy and Efficient Finetuning LLMs. (Supported LLama, LLama2, LLama3, Qwen, Baichuan, GLM , Falcon) 大模型高效量化训练+部署.
https://jianzhnie.github.io/llmtech/
Apache License 2.0
561 stars 61 forks source link

RuntimeError: probability tensor contains either `inf`, `nan` or element < 0 #5

Closed weifan-zhao closed 1 year ago

weifan-zhao commented 1 year ago
截屏2023-06-13 09 29 19
jianzhnie commented 1 year ago

which model do you run? the base_model and adapters

weifan-zhao commented 1 year ago

which model do you run? the base_model and adapters

the base model: vicuna-13b adapters: trained with finetune.sh(dataset_format: alpaca-cleaned)

jianzhnie commented 1 year ago
python generate_server.py \
    --model_name_or_path decapoda-research/llama-7b-hf \
    --lora_model_name_or_path  tloen/alpaca-lora-7b \
    --load_8bit

run add test this scripts and find the bugs

weifan-zhao commented 1 year ago
python generate_server.py \
    --model_name_or_path decapoda-research/llama-7b-hf \
    --lora_model_name_or_path  tloen/alpaca-lora-7b \
    --load_8bit

run add test this scripts and find the bugs

I ran this script with the same error as before