关于GCU、沐曦GPGPU、MLU、0卡V100资源4月7日恢复上架的公告>>> 关于共建具身智能开源数据集的倡议>>> 关于云脑任务中统一路径访问方式的公告>>> 关于将启智集群GPU资源迁移至智算集群的公告>>>
thomas-yanxin
/
PaddleNLP
mirror of https://github.com/PaddlePaddle/PaddleNLP.git
Not watched
1
0
Fork 0
Code
Releases 47 Wiki Activity
Issues 0 Datasets Model Cloudbrain HPC
You can not select more than 25 topics Topics must start with a chinese character,a letter or number, can include dashes ('-') and can be up to 35 characters long.
4798 Commits
34 Branches
590 MiB
304 Download
 
 
 
 
 
 
Branch: release/2.8
Branches Tags
${ item.name }
Create branch ${ searchTerm }
from 'release/2.8'
${ noResults }
PaddleNLP/llm/opt
Ting Liu 445638a4ab
llm doc update (#7478) 		5 months ago
..
README.md [llm]support get last checkpoint (#6914) 8 months ago
lora_argument.json llm doc update (#7478) 5 months ago
sft_argument.json llm doc update (#7478) 5 months ago

README.md

OPT

1. 模型介绍

OPT: Open Pre-trained Transformer Language Models 是以自回归填空作为训练目标的通用语言模型，可用于各类理解和生成任务。

支持模型权重:

Model
facebook/opt-125m
facebook/opt-350m
facebook/opt-1.3b
facebook/opt-2.7b
facebook/opt-6.7b
facebook/opt-13b
facebook/opt-30b
facebook/opt-66b
facebook/opt-iml-1.3b
opt-iml-max-1.3b

2. 模型精调

请参考LLM全流程工具介绍

Delete task

Are you sure you want to delete this task? Once this task is deleted, it cannot be recovered.

cancel
confirm

No Description

Python C++ Shell Cuda Markdown other

Please read the following content carefully:

Dear OpenI User

Thank you for your continuous support to the Openl Qizhi Community AI Collaboration Platform. In order to protect your usage rights and ensure network security, we updated the Openl Qizhi Community AI Collaboration Platform Usage Agreement in January 2024. The updated agreement specifies that users are prohibited from using intranet penetration tools. After you click "Agree and continue", you can continue to use our services. Thank you for your cooperation and understanding.

For more agreement content, please refer to the《Openl Qizhi Community AI Collaboration Platform Usage Agreement》

Agree and continue
Disagree, exit