-
如果你遇到问题需要我们帮助,你可以从以下角度描述你的信息,以便于我们可以理解或者复现你的错误(学会如何提问不仅是能帮助我们理解你,也是一个自查的过程):
1、你使用了哪个脚本、使用的什么命令
`bash finetune_continue.sh`
2、你的参数是什么(脚本参数、命令参数)
```
TOT_CUDA="0,1"
CUDAs=(${TOT_CUDA//,/ })
CUDA…
-
如果你遇到问题需要我们帮助,你可以从以下角度描述你的信息,以便于我们可以理解或者复现你的错误(学会如何提问不仅是能帮助我们理解你,也是一个自查的过程):
1、你使用了哪个脚本、使用的什么命令
bash finetune.sh
2、你的参数是什么(脚本参数、命令参数)
TOT_CUDA="0"
CUDAs=(${TOT_CUDA//,/ })
CUDA_NUM=${#CUD…
-
```
Successfully built transformers
Installing collected packages: transformers
Successfully installed transformers-4.28.0.dev0
WARNING: Running pip as the 'root' user can result in broken permiss…
-
使用finetune_continue脚本,2080ti单卡,跑了120个小时(跑到17298了。),出来的adapter_model.bin,使用generate.sh脚本加载进行问答,基本不能用。感觉这刚开始炼丹还不是一般的难。看着大神的医疗问答样例,感觉差距太遥远了,深刻求指导。
看了下checkpoint-17200里边的trainer_state.json,发现loss是一直有降,…
-
@PanQiWei @TheBloke
So Here is my Script for Infrence:
```
import torch
from transformers import AutoTokenizer, TextGenerationPipeline
from auto_gptq import AutoGPTQForCausalLM
from huggingf…
-
输出如下
Epoch 04383: adjusting learning rate of group 0 to 9.9171e-06.
Epoch 04383: adjusting learning rate of group 1 to 9.9171e-06.
Epoch 04383: adjusting learning rate of group 0 to 9.9171e-06.
Ep…
-
运行chat.sh启动服务之后,在生成结果的时候,显存使用会持续增加,直至最后显存溢出,这个有什么解决方法吗
-
### Describe the bug
Auto-GPT can't convert any output from the API to JSON
Settings I've tried:
- python server.py --chat --listen --xformers --monkey-patch --bf16
- python server.py --chat -…
-
-
# Airport Name
Pampa Guanaco
# Country?
Chile
# Improvements that need to be made?
Add airport to the flight sim
# Are you working on this airport?
Yes
# Airport Priority? (10000ft+ Runway, Lo…