René's URL Explorer Experiment


Title: GitHub - flamecodezz/llm-action: 本项目旨在分享大模型相关技术原理以及实战经验(大模型工程化、大模型应用落地) · GitHub

Open Graph Title: GitHub - flamecodezz/llm-action: 本项目旨在分享大模型相关技术原理以及实战经验(大模型工程化、大模型应用落地)

X Title: GitHub - flamecodezz/llm-action: 本项目旨在分享大模型相关技术原理以及实战经验(大模型工程化、大模型应用落地)

Description: 本项目旨在分享大模型相关技术原理以及实战经验(大模型工程化、大模型应用落地). Contribute to flamecodezz/llm-action development by creating an account on GitHub.

Open Graph Description: 本项目旨在分享大模型相关技术原理以及实战经验(大模型工程化、大模型应用落地). Contribute to flamecodezz/llm-action development by creating an account on GitHub.

X Description: 本项目旨在分享大模型相关技术原理以及实战经验(大模型工程化、大模型应用落地). Contribute to flamecodezz/llm-action development by creating an account on GitHub.

Opengraph URL: https://github.com/flamecodezz/llm-action

X: @github

direct link

Domain: github.com

IPv4 address: 140.82.114.3

react-profiling0
react-import-mapreact,react-dom,react-dom/client,react-dom/profiling,react-is,react/compiler-runtime,react/jsx-dev-runtime,react/jsx-runtime,scheduler@b8488ee2c322
route-pattern/:user_id/:repository
route-controllerfiles
route-actiondisambiguate
fetch-noncev2:65e2f3f7-21e0-0b24-cd08-4fa0735a27cc
current-catalog-service-hashf3abb0cc802f3d7b95fc8762b94bdcb13bf39634c40c357301c4aa1d67a256fb
request-idB10A:D500:265B099:32A36A4:6B037450
html-safe-nonce613d21d1aa21da9848e8109ed1c4a8230f32ef5003158fee227a882a379962fa
visitor-payloadeyJyZWZlcnJlciI6IiIsInJlcXVlc3RfaWQiOiJCMTBBOkQ1MDA6MjY1QjA5OTozMkEzNkE0OjZCMDM3NDUwIiwidmlzaXRvcl9pZCI6IjIyODgyMDE1NzQ1NzQ2ODMyMTYiLCJyZWdpb25fZWRnZSI6ImlhZCIsInJlZ2lvbl9yZW5kZXIiOiJpYWQifQ==
visitor-hmac1c701d41abdbc4a69f5256a06252d083efcf66775f90c7ff955b4cbcf20fd8c9
hovercard-subject-tagrepository:901269528
github-keyboard-shortcutsrepository,copilot
google-site-verificationApib7-x98H0j5cPqHWwSMm6dNU4GmODRoqxLiDzdx9I
octolytics-urlhttps://collector.github.com/github/collect
analytics-location//
fb:app_id1401488693436528
apple-itunes-appapp-id=1477376905, app-argument=https://github.com/flamecodezz/llm-action
twitter:imagehttps://opengraph.githubassets.com/3d321a217289144fbf8d2f94c28228a684555fa1e13c78249317889806f672e3/flamecodezz/llm-action
twitter:cardsummary_large_image
og:imagehttps://opengraph.githubassets.com/3d321a217289144fbf8d2f94c28228a684555fa1e13c78249317889806f672e3/flamecodezz/llm-action
og:image:alt本项目旨在分享大模型相关技术原理以及实战经验(大模型工程化、大模型应用落地). Contribute to flamecodezz/llm-action development by creating an account on GitHub.
og:image:width1200
og:image:height600
og:site_nameGitHub
og:typeobject
hostnamegithub.com
expected-hostnamegithub.com
None87e774ee7c6b3bee0490394e4d3e57ec9dbc956f74b87ea685dc8996c1676456
turbo-cache-controlno-cache
go-importgithub.com/flamecodezz/llm-action git https://github.com/flamecodezz/llm-action.git
octolytics-dimension-user_id48575118
octolytics-dimension-user_loginflamecodezz
octolytics-dimension-repository_id901269528
octolytics-dimension-repository_nwoflamecodezz/llm-action
octolytics-dimension-repository_publictrue
octolytics-dimension-repository_is_forktrue
octolytics-dimension-repository_parent_id644235905
octolytics-dimension-repository_parent_nwoliguodongiot/llm-action
octolytics-dimension-repository_network_root_id644235905
octolytics-dimension-repository_network_root_nwoliguodongiot/llm-action
turbo-body-classeslogged-out env-production page-responsive
disable-turbofalse
browser-stats-urlhttps://api.github.com/_private/browser/stats
browser-errors-urlhttps://api.github.com/_private/browser/errors
release423dbb54c5e7bfecc838eb387c1435eb3d0e760d
ui-targetfull
theme-color#1e2327
color-schemelight dark

Links:

Skip to contenthttps://github.com/flamecodezz/llm-action#start-of-content
https://github.com/
Sign inhttps://github.com/login?return_to=https%3A%2F%2Fgithub.com%2Fflamecodezz%2Fllm-action
GitHubRepository hosting, code review, and collaborationhttps://github.com/platform
CopilotAI coding agent for tasks, projects, and automated workhttps://github.com/features/copilot
ActionsCI/CD and automationhttps://github.com/features/actions
Code QualityQuality analysis, fixes, and enforcementhttps://github.com/features/code-quality
CodespacesCloud development environmentshttps://github.com/features/codespaces
Code SecurityVulnerability detection and remediationhttps://github.com/security/advanced-security/code-security
Secret ProtectionSecret scanning and leak preventionhttps://github.com/security/advanced-security/secret-protection
All Featureshttps://github.com/features
Documentationhttps://docs.github.com/en
Traininghttps://learn.github.com/skills
Certificationshttps://learn.github.com/credentials
Build with Copilothttps://github.com/features/ai
Customize Copilothttps://docs.github.com/en/copilot/how-tos/copilot-on-github/customize-copilot/customize-copilot-overview
Integrationshttps://github.com/marketplace
Trending reposhttps://github.com/trending
GitHub Sponsorshttps://github.com/open-source/sponsors
Developer articleshttps://github.com/resources/articles
ChangelogYour source for the latest features, improvements, and fixes across GitHub products, all in one place.https://github.blog/changelog/
Enterpriseshttps://github.com/enterprise
Teamhttps://github.com/team
Startupshttps://github.com/enterprise/startups
Nonprofitshttps://github.com/solutions/industry/nonprofits
Healthcarehttps://github.com/solutions/industry/healthcare
Financial serviceshttps://github.com/solutions/industry/financial-services
Manufacturinghttps://github.com/solutions/industry/manufacturing
Governmenthttps://github.com/solutions/industry/government
See all industrieshttps://github.com/solutions/industry
Why GitHubhttps://github.com/why-github
Customer Storieshttps://github.com/customer-stories
Business insightshttps://github.com/solutions/executive-insights
Trust Centerhttps://github.com/trust-center
GitHub Advanced SecurityEnterprise-grade code securityhttps://github.com/security/advanced-security
Copilot for businessEnterprise-grade AI featureshttps://github.com/features/copilot/copilot-business
Premium SupportEnterprise-grade 24/7 supporthttps://github.com/enterprise/premium-support
Documentationhttps://docs.github.com/en
Resource centerhttps://github.com/resources
Bloghttps://github.blog/
Events & webinarshttps://github.com/resources/events
Community Forumhttps://github.com/orgs/community/discussions
Customer Storieshttps://github.com/customer-stories
Integrationshttps://github.com/marketplace
Customer supporthttps://help.github.com/
Traininghttps://learn.github.com/skills
GitHub Statushttps://www.githubstatus.com/
Abouthttps://github.com/about
Partnershttps://github.com/partners
Changeloghttps://github.blog/changelog/
GitHub UniverseThe event for developers, by developers — talks, workshops, and what’s next for GitHub.https://githubuniverse.com/
Pricinghttps://github.com/pricing
Sign inhttps://github.com/login?return_to=https%3A%2F%2Fgithub.com%2Fflamecodezz%2Fllm-action
Sign uphttps://github.com/signup?ref_cta=Sign+up&ref_loc=header+logged+out&ref_page=%2F%3Cuser-name%3E%2F%3Crepo-name%3E&source=header-repo&source_repo=flamecodezz%2Fllm-action
Reloadhttps://github.com/flamecodezz/llm-action
Reloadhttps://github.com/flamecodezz/llm-action
Reloadhttps://github.com/flamecodezz/llm-action
flamecodezzhttps://github.com/flamecodezz
llm-actionhttps://github.com/flamecodezz/llm-action
Codehttps://github.com/flamecodezz/llm-action
Pull requestshttps://github.com/flamecodezz/llm-action/pulls
Actionshttps://github.com/flamecodezz/llm-action/actions
Projectshttps://github.com/flamecodezz/llm-action/projects
Security and qualityhttps://github.com/flamecodezz/llm-action/security
Insightshttps://github.com/flamecodezz/llm-action/pulse
https://github.com/flamecodezz/llm-action
llm-actionhttps://github.com/flamecodezz/llm-action
liguodongiot/llm-actionhttps://github.com/liguodongiot/llm-action
Notificationshttps://github.com/login?return_to=%2Fflamecodezz%2Fllm-action
Fork0 (0)https://github.com/login?return_to=%2Fflamecodezz%2Fllm-action
Star1 (1)https://github.com/login?return_to=%2Fflamecodezz%2Fllm-action
www.zhihu.com/column/c_1456193767213043713https://www.zhihu.com/column/c_1456193767213043713
Readmehttps://github.com/flamecodezz/llm-action#readme-ov-file
Licensehttps://github.com/flamecodezz/llm-action#License-1-ov-file
Activityhttps://github.com/flamecodezz/llm-action/activity
0 forkshttps://github.com/flamecodezz/llm-action/forks
Report repositoryhttps://github.com/contact/report-content?content_url=https%3A%2F%2Fgithub.com%2Fflamecodezz%2Fllm-action&report=flamecodezz+%28user%29
Brancheshttps://github.com/flamecodezz/llm-action/branches
Tagshttps://github.com/flamecodezz/llm-action/tags
https://github.com/flamecodezz/llm-action/branches
https://github.com/flamecodezz/llm-action/tags
899 Commitshttps://github.com/flamecodezz/llm-action/commits/main/
https://github.com/flamecodezz/llm-action/commits/main/
ai-compilerhttps://github.com/flamecodezz/llm-action/tree/main/ai-compiler
ai-compilerhttps://github.com/flamecodezz/llm-action/tree/main/ai-compiler
ai-frameworkhttps://github.com/flamecodezz/llm-action/tree/main/ai-framework
ai-frameworkhttps://github.com/flamecodezz/llm-action/tree/main/ai-framework
ai-infrahttps://github.com/flamecodezz/llm-action/tree/main/ai-infra
ai-infrahttps://github.com/flamecodezz/llm-action/tree/main/ai-infra
bloghttps://github.com/flamecodezz/llm-action/tree/main/blog
bloghttps://github.com/flamecodezz/llm-action/tree/main/blog
docshttps://github.com/flamecodezz/llm-action/tree/main/docs
docshttps://github.com/flamecodezz/llm-action/tree/main/docs
faqhttps://github.com/flamecodezz/llm-action/tree/main/faq
faqhttps://github.com/flamecodezz/llm-action/tree/main/faq
llm-algohttps://github.com/flamecodezz/llm-action/tree/main/llm-algo
llm-algohttps://github.com/flamecodezz/llm-action/tree/main/llm-algo
llm-alignmenthttps://github.com/flamecodezz/llm-action/tree/main/llm-alignment
llm-alignmenthttps://github.com/flamecodezz/llm-action/tree/main/llm-alignment
llm-applicationhttps://github.com/flamecodezz/llm-action/tree/main/llm-application
llm-applicationhttps://github.com/flamecodezz/llm-action/tree/main/llm-application
llm-benchmarkhttps://github.com/flamecodezz/llm-action/tree/main/llm-benchmark
llm-benchmarkhttps://github.com/flamecodezz/llm-action/tree/main/llm-benchmark
llm-compressionhttps://github.com/flamecodezz/llm-action/tree/main/llm-compression
llm-compressionhttps://github.com/flamecodezz/llm-action/tree/main/llm-compression
llm-data-engineeringhttps://github.com/flamecodezz/llm-action/tree/main/llm-data-engineering
llm-data-engineeringhttps://github.com/flamecodezz/llm-action/tree/main/llm-data-engineering
llm-inferencehttps://github.com/flamecodezz/llm-action/tree/main/llm-inference
llm-inferencehttps://github.com/flamecodezz/llm-action/tree/main/llm-inference
llm-interviewhttps://github.com/flamecodezz/llm-action/tree/main/llm-interview
llm-interviewhttps://github.com/flamecodezz/llm-action/tree/main/llm-interview
llm-localizationhttps://github.com/flamecodezz/llm-action/tree/main/llm-localization
llm-localizationhttps://github.com/flamecodezz/llm-action/tree/main/llm-localization
llm-maashttps://github.com/flamecodezz/llm-action/tree/main/llm-maas
llm-maashttps://github.com/flamecodezz/llm-action/tree/main/llm-maas
llm-optimizerhttps://github.com/flamecodezz/llm-action/tree/main/llm-optimizer
llm-optimizerhttps://github.com/flamecodezz/llm-action/tree/main/llm-optimizer
llm-performancehttps://github.com/flamecodezz/llm-action/tree/main/llm-performance
llm-performancehttps://github.com/flamecodezz/llm-action/tree/main/llm-performance
llm-pipelinehttps://github.com/flamecodezz/llm-action/tree/main/llm-pipeline
llm-pipelinehttps://github.com/flamecodezz/llm-action/tree/main/llm-pipeline
llm-trainhttps://github.com/flamecodezz/llm-action/tree/main/llm-train
llm-trainhttps://github.com/flamecodezz/llm-action/tree/main/llm-train
llmopshttps://github.com/flamecodezz/llm-action/tree/main/llmops
llmopshttps://github.com/flamecodezz/llm-action/tree/main/llmops
paperhttps://github.com/flamecodezz/llm-action/tree/main/paper
paperhttps://github.com/flamecodezz/llm-action/tree/main/paper
pichttps://github.com/flamecodezz/llm-action/tree/main/pic
pichttps://github.com/flamecodezz/llm-action/tree/main/pic
.DS_Storehttps://github.com/flamecodezz/llm-action/blob/main/.DS_Store
.DS_Storehttps://github.com/flamecodezz/llm-action/blob/main/.DS_Store
.gitignorehttps://github.com/flamecodezz/llm-action/blob/main/.gitignore
.gitignorehttps://github.com/flamecodezz/llm-action/blob/main/.gitignore
LICENSEhttps://github.com/flamecodezz/llm-action/blob/main/LICENSE
LICENSEhttps://github.com/flamecodezz/llm-action/blob/main/LICENSE
README.mdhttps://github.com/flamecodezz/llm-action/blob/main/README.md
README.mdhttps://github.com/flamecodezz/llm-action/blob/main/README.md
git-pull-push.shhttps://github.com/flamecodezz/llm-action/blob/main/git-pull-push.sh
git-pull-push.shhttps://github.com/flamecodezz/llm-action/blob/main/git-pull-push.sh
mkdir-dir-file.shhttps://github.com/flamecodezz/llm-action/blob/main/mkdir-dir-file.sh
mkdir-dir-file.shhttps://github.com/flamecodezz/llm-action/blob/main/mkdir-dir-file.sh
READMEhttps://github.com/flamecodezz/llm-action
Licensehttps://github.com/flamecodezz/llm-action
https://github.com/liguodongiot/llm-action/blob/main/pic/llm-action-v3.png
https://github.com/liguodongiot/llm-action/stargazers
https://github.com/liguodongiot/llm-action/blob/main/pic/wx.jpg
https://www.zhihu.com/people/liguodong-iot
https://juejin.cn/user/3642056016410728
https://liguodong.blog.csdn.net/
unify-easy-llmhttps://github.com/liguodongiot/unify-easy-llm
https://github.com/flamecodezz/llm-action#目录
LLM训练https://github.com/flamecodezz/llm-action#llm%E8%AE%AD%E7%BB%83
LLM训练实战https://github.com/flamecodezz/llm-action#llm%E8%AE%AD%E7%BB%83%E5%AE%9E%E6%88%98
LLM参数高效微调技术原理https://github.com/flamecodezz/llm-action#llm%E5%BE%AE%E8%B0%83%E6%8A%80%E6%9C%AF%E5%8E%9F%E7%90%86
LLM参数高效微调技术实战https://github.com/flamecodezz/llm-action#llm%E5%BE%AE%E8%B0%83%E5%AE%9E%E6%88%98
LLM分布式训练并行技术https://github.com/flamecodezz/llm-action#llm%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83%E5%B9%B6%E8%A1%8C%E6%8A%80%E6%9C%AF
分布式AI框架https://github.com/flamecodezz/llm-action#%E5%88%86%E5%B8%83%E5%BC%8Fai%E6%A1%86%E6%9E%B6
分布式训练网络通信https://github.com/flamecodezz/llm-action#%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83%E7%BD%91%E7%BB%9C%E9%80%9A%E4%BF%A1
LLM训练优化技术https://github.com/flamecodezz/llm-action#llm%E8%AE%AD%E7%BB%83%E4%BC%98%E5%8C%96%E6%8A%80%E6%9C%AF
LLM对齐技术https://github.com/flamecodezz/llm-action#llm%E5%AF%B9%E9%BD%90%E6%8A%80%E6%9C%AF
LLM推理https://github.com/flamecodezz/llm-action#llm%E6%8E%A8%E7%90%86
LLM推理框架https://github.com/flamecodezz/llm-action#llm%E6%8E%A8%E7%90%86%E6%A1%86%E6%9E%B6
LLM推理优化技术https://github.com/flamecodezz/llm-action#llm%E6%8E%A8%E7%90%86%E4%BC%98%E5%8C%96%E6%8A%80%E6%9C%AF
LLM压缩https://github.com/flamecodezz/llm-action#llm%E5%8E%8B%E7%BC%A9
LLM量化https://github.com/flamecodezz/llm-action#llm%E9%87%8F%E5%8C%96
LLM剪枝https://github.com/flamecodezz/llm-action#llm%E5%89%AA%E6%9E%9D
LLM知识蒸馏https://github.com/flamecodezz/llm-action#llm%E7%9F%A5%E8%AF%86%E8%92%B8%E9%A6%8F
低秩分解https://github.com/flamecodezz/llm-action#%E4%BD%8E%E7%A7%A9%E5%88%86%E8%A7%A3
LLM测评https://github.com/flamecodezz/llm-action#llm%E6%B5%8B%E8%AF%84
LLM数据工程https://github.com/flamecodezz/llm-action#llm%E6%95%B0%E6%8D%AE%E5%B7%A5%E7%A8%8B
LLM微调高效数据筛选技术https://github.com/flamecodezz/llm-action#llm%E5%BE%AE%E8%B0%83%E9%AB%98%E6%95%88%E6%95%B0%E6%8D%AE%E7%AD%9B%E9%80%89%E6%8A%80%E6%9C%AF
提示工程https://github.com/flamecodezz/llm-action#%E6%8F%90%E7%A4%BA%E5%B7%A5%E7%A8%8B
LLM算法架构https://github.com/flamecodezz/llm-action#llm%E7%AE%97%E6%B3%95%E6%9E%B6%E6%9E%84
LLM应用开发https://github.com/flamecodezz/llm-action#llm%E5%BA%94%E7%94%A8%E5%BC%80%E5%8F%91
LLM国产化适配https://github.com/flamecodezz/llm-action#llm%E5%9B%BD%E4%BA%A7%E5%8C%96%E9%80%82%E9%85%8D
AI编译器https://github.com/flamecodezz/llm-action#ai%E7%BC%96%E8%AF%91%E5%99%A8
AI基础设施https://github.com/flamecodezz/llm-action#ai%E5%9F%BA%E7%A1%80%E8%AE%BE%E6%96%BD
AI加速卡https://github.com/flamecodezz/llm-action#ai%E5%8A%A0%E9%80%9F%E5%8D%A1
AI集群网络通信https://github.com/flamecodezz/llm-action#ai%E9%9B%86%E7%BE%A4%E7%BD%91%E7%BB%9C%E9%80%9A%E4%BF%A1
LLMOpshttps://github.com/flamecodezz/llm-action#llmops
LLM生态相关技术https://github.com/flamecodezz/llm-action#llm%E7%94%9F%E6%80%81%E7%9B%B8%E5%85%B3%E6%8A%80%E6%9C%AF
LLM面试题https://github.com/flamecodezz/llm-action#llm%E9%9D%A2%E8%AF%95%E9%A2%98
服务器基础环境安装及常用工具https://github.com/flamecodezz/llm-action#%E6%9C%8D%E5%8A%A1%E5%99%A8%E5%9F%BA%E7%A1%80%E7%8E%AF%E5%A2%83%E5%AE%89%E8%A3%85%E5%8F%8A%E5%B8%B8%E7%94%A8%E5%B7%A5%E5%85%B7
LLM学习交流群https://github.com/flamecodezz/llm-action#llm%E5%AD%A6%E4%B9%A0%E4%BA%A4%E6%B5%81%E7%BE%A4
微信公众号https://github.com/flamecodezz/llm-action#%E5%BE%AE%E4%BF%A1%E5%85%AC%E4%BC%97%E5%8F%B7
Star Historyhttps://github.com/flamecodezz/llm-action#star-history
AI工程化课程推荐https://github.com/flamecodezz/llm-action#ai%E5%B7%A5%E7%A8%8B%E5%8C%96%E8%AF%BE%E7%A8%8B%E6%8E%A8%E8%8D%90
https://github.com/flamecodezz/llm-action#llm训练
https://github.com/flamecodezz/llm-action#llm训练实战
从0到1复现斯坦福羊驼(Stanford Alpaca 7B)https://zhuanlan.zhihu.com/p/618321077
配套代码https://github.com/liguodongiot/llm-action/tree/main/llm-train/alpaca
足够惊艳,使用Alpaca-Lora基于LLaMA(7B)二十分钟完成微调,效果比肩斯坦福羊驼https://zhuanlan.zhihu.com/p/619426866
使用 LoRA 技术对 LLaMA 65B 大模型进行微调及推理https://zhuanlan.zhihu.com/p/632492604
配套代码https://github.com/liguodongiot/llm-action/tree/main/llm-train/alpaca-lora
基于LLaMA-7B/Bloomz-7B1-mt复现开源中文对话大模型BELLE及GPTQ量化https://zhuanlan.zhihu.com/p/618876472
BELLE(LLaMA-7B/Bloomz-7B1-mt)大模型使用GPTQ量化后推理性能测试https://zhuanlan.zhihu.com/p/621128368
从0到1基于ChatGLM-6B使用LoRA进行参数高效微调https://zhuanlan.zhihu.com/p/621793987
配套代码https://github.com/liguodongiot/llm-action/tree/main/train/chatglm-lora
使用DeepSpeed/P-Tuning v2对ChatGLM-6B进行微调https://zhuanlan.zhihu.com/p/622351059
配套代码https://github.com/liguodongiot/llm-action/tree/main/train/chatglm
大模型也内卷,Vicuna训练及推理指南,效果碾压斯坦福羊驼https://zhuanlan.zhihu.com/p/624012908
一键式 RLHF 训练 DeepSpeed Chat(一):理论篇https://zhuanlan.zhihu.com/p/626159553
一键式 RLHF 训练 DeepSpeed Chat(二):实践篇https://zhuanlan.zhihu.com/p/626214655
配套代码https://github.com/liguodongiot/llm-action/tree/main/train/deepspeedchat
大杀器,多模态大模型MiniGPT-4入坑指南https://zhuanlan.zhihu.com/p/627671257
中文LLaMA&Alpaca大语言模型词表扩充+预训练+指令精调https://zhuanlan.zhihu.com/p/631360711
配套代码https://github.com/liguodongiot/llm-action/tree/main/train/chinese-llama-alpaca
高效微调技术QLoRA实战,基于LLaMA-65B微调仅需48G显存,真香https://zhuanlan.zhihu.com/p/636644164
配套代码https://github.com/liguodongiot/llm-action/tree/main/train/qlora
突破内存瓶颈,使用 GaLore 一张4090消费级显卡也能预训练LLaMA-7Bhttps://zhuanlan.zhihu.com/p/686686751
配套代码https://github.com/liguodongiot/llm-action/blob/main/train/galore/torchrun_main.py
⬆ 一键返回目录https://github.com/flamecodezz/llm-action#%E7%9B%AE%E5%BD%95
https://github.com/flamecodezz/llm-action#llm微调技术原理
https://github.com/flamecodezz/llm-action/blob/main/pic/llm/train/sft/peft%E6%96%B9%E6%B3%95.jpg
大模型参数高效微调技术原理综述(一)-背景、参数高效微调简介https://zhuanlan.zhihu.com/p/635152813
大模型参数高效微调技术原理综述(二)-BitFit、Prefix Tuning、Prompt Tuninghttps://zhuanlan.zhihu.com/p/635686756
大模型参数高效微调技术原理综述(三)-P-Tuning、P-Tuning v2https://zhuanlan.zhihu.com/p/635848732
大模型参数高效微调技术原理综述(四)-Adapter Tuning及其变体https://zhuanlan.zhihu.com/p/636038478
大模型参数高效微调技术原理综述(五)-LoRA、AdaLoRA、QLoRAhttps://zhuanlan.zhihu.com/p/636215898
大模型参数高效微调技术原理综述(六)-MAM Adapter、UniPELThttps://zhuanlan.zhihu.com/p/636362246
大模型参数高效微调技术原理综述(七)-最佳实践、总结https://zhuanlan.zhihu.com/p/649755252
https://github.com/flamecodezz/llm-action#llm微调实战
大模型参数高效微调技术实战(一)-PEFT概述及环境搭建https://zhuanlan.zhihu.com/p/651744834
大模型参数高效微调技术实战(二)-Prompt Tuninghttps://zhuanlan.zhihu.com/p/646748939
配套代码https://github.com/liguodongiot/llm-action/blob/main/llm-action/peft/clm/peft_prompt_tuning_clm.ipynb
大模型参数高效微调技术实战(三)-P-Tuninghttps://zhuanlan.zhihu.com/p/646876256
配套代码https://github.com/liguodongiot/llm-action/blob/main/llm-action/peft/clm/peft_p_tuning_clm.ipynb
大模型参数高效微调技术实战(四)-Prefix Tuning / P-Tuning v2https://zhuanlan.zhihu.com/p/648156780
配套代码https://github.com/liguodongiot/llm-action/blob/main/llm-action/peft/clm/peft_p_tuning_v2_clm.ipynb
大模型参数高效微调技术实战(五)-LoRAhttps://zhuanlan.zhihu.com/p/649315197
配套代码https://github.com/liguodongiot/llm-action/blob/main/llm-action/peft/clm/peft_lora_clm.ipynb
大模型参数高效微调技术实战(六)-IA3https://zhuanlan.zhihu.com/p/649707359
配套代码https://github.com/liguodongiot/llm-action/blob/main/llm-action/peft/clm/peft_ia3_clm.ipynb
大模型微调实战(七)-基于LoRA微调多模态大模型https://zhuanlan.zhihu.com/p/670048482
配套代码https://github.com/liguodongiot/llm-action/blob/main/llm-action/peft/multimodal/blip2_lora_int8_fine_tune.py
大模型微调实战(八)-使用INT8/FP4/NF4微调大模型https://zhuanlan.zhihu.com/p/670116171
配套代码https://github.com/liguodongiot/llm-action/blob/main/llm-action/peft/multimodal/finetune_bloom_bnb_peft.ipynb
⬆ 一键返回目录https://github.com/flamecodezz/llm-action#%E7%9B%AE%E5%BD%95
LLM分布式训练并行技术https://github.com/liguodongiot/llm-action/tree/main/docs/llm-base/distribution-parallelism
https://github.com/flamecodezz/llm-action#llm分布式训练并行技术
大模型分布式训练并行技术(一)-概述https://zhuanlan.zhihu.com/p/598714869
大模型分布式训练并行技术(二)-数据并行https://zhuanlan.zhihu.com/p/650002268
大模型分布式训练并行技术(三)-流水线并行https://zhuanlan.zhihu.com/p/653860567
大模型分布式训练并行技术(四)-张量并行https://zhuanlan.zhihu.com/p/657921100
大模型分布式训练并行技术(五)-序列并行https://zhuanlan.zhihu.com/p/659792351
大模型分布式训练并行技术(六)-多维混合并行https://zhuanlan.zhihu.com/p/661279318
大模型分布式训练并行技术(七)-自动并行https://zhuanlan.zhihu.com/p/662517647
大模型分布式训练并行技术(八)-MOE并行https://zhuanlan.zhihu.com/p/662518387
大模型分布式训练并行技术(九)-总结https://zhuanlan.zhihu.com/p/667051845
⬆ 一键返回目录https://github.com/flamecodezz/llm-action#%E7%9B%AE%E5%BD%95
https://github.com/flamecodezz/llm-action#分布式ai框架
PyTorchhttps://github.com/liguodongiot/llm-action/tree/main/train/pytorch/
Megatron-LMhttps://github.com/liguodongiot/llm-action/tree/main/train/megatron
基于Megatron-LM从0到1完成GPT2模型预训练、模型评估及推理https://juejin.cn/post/7259682893648724029
DeepSpeedhttps://github.com/liguodongiot/llm-action/tree/main/train/deepspeed
Megatron-DeepSpeedhttps://github.com/liguodongiot/llm-action/tree/main/train/megatron-deepspeed
https://github.com/flamecodezz/llm-action#分布式训练网络通信
https://github.com/flamecodezz/llm-action#llm训练优化技术
https://github.com/flamecodezz/llm-action#llm对齐技术
⬆ 一键返回目录https://github.com/flamecodezz/llm-action#%E7%9B%AE%E5%BD%95
LLM推理https://github.com/liguodongiot/llm-action/tree/main/inference
https://github.com/flamecodezz/llm-action#llm推理
https://github.com/flamecodezz/llm-action#llm推理框架
大模型推理框架概述https://www.zhihu.com/question/625415776/answer/3243562246
大模型的好伙伴,浅析推理加速引擎FasterTransformerhttps://zhuanlan.zhihu.com/p/626008090
模型推理服务化框架Triton保姆式教程(一):快速入门https://zhuanlan.zhihu.com/p/629336492
模型推理服务化框架Triton保姆式教程(二):架构解析https://zhuanlan.zhihu.com/p/634143650
模型推理服务化框架Triton保姆式教程(三):开发实践https://zhuanlan.zhihu.com/p/634444666
TensorRT-LLM保姆级教程(一)-快速入门https://zhuanlan.zhihu.com/p/666849728
TensorRT-LLM保姆级教程(二)-离线环境搭建、模型量化及推理https://zhuanlan.zhihu.com/p/667572720
TensorRT-LLM保姆级教程(三)-使用Triton推理服务框架部署模型https://juejin.cn/post/7398122968200593419
https://github.com/flamecodezz/llm-action#llm推理优化技术
LLM推理优化技术-概述https://github.com/flamecodezz/llm-action/blob/main
大模型推理优化技术-KV Cachehttps://www.zhihu.com/question/653658936/answer/3569365986
大模型推理服务调度优化技术-Continuous batchinghttps://zhuanlan.zhihu.com/p/719610083
大模型低显存推理优化-Offload技术https://juejin.cn/post/7405158045628596224
大模型推理优化技术-KV Cache量化https://juejin.cn/post/7420231738558627874
大模型推理优化技术-KV Cache优化方法综述https://github.com/flamecodezz/llm-action/blob/main
https://github.com/flamecodezz/llm-action#llm压缩
LLM量化https://github.com/liguodongiot/llm-action/tree/main/model-compression/quantization
https://github.com/flamecodezz/llm-action#llm量化
大模型量化概述https://www.zhihu.com/question/627484732/answer/3261671478
大模型量化感知训练技术原理:LLM-QAThttps://zhuanlan.zhihu.com/p/647589650
大模型量化感知微调技术原理:QLoRAhttps://github.com/flamecodezz/llm-action/blob/main
大模型量化技术原理:GPTQ、LLM.int8()https://zhuanlan.zhihu.com/p/680212402
大模型量化技术原理:SmoothQuanthttps://www.zhihu.com/question/576376372/answer/3388402085
大模型量化技术原理:AWQ、AutoAWQhttps://zhuanlan.zhihu.com/p/681578090
大模型量化技术原理:SpQRhttps://zhuanlan.zhihu.com/p/682871823
大模型量化技术原理:ZeroQuant系列https://zhuanlan.zhihu.com/p/683813769
大模型量化技术原理:FP8https://www.zhihu.com/question/658712811/answer/3596678896
大模型量化技术原理:FP6https://juejin.cn/post/7412893752090853386
大模型量化技术原理:KIVI、IntactKV、KVQuanthttps://juejin.cn/post/7420231738558627874
大模型量化技术原理:Atom、QuaRothttps://juejin.cn/post/7424334647570513972
大模型量化技术原理:QoQ量化及QServe推理服务系统https://github.com/flamecodezz/llm-action/blob/main
大模型量化技术原理:FP4https://github.com/flamecodezz/llm-action/blob/main
大模型量化技术原理:总结https://github.com/flamecodezz/llm-action/blob/main
https://github.com/flamecodezz/llm-action#llm剪枝
万字长文谈深度神经网络剪枝综述https://zhuanlan.zhihu.com/p/692858636
大模型剪枝技术原理:概述https://www.zhihu.com/question/652126515/answer/3457652467
大模型剪枝技术原理:LLM-Pruner、SliceGPThttps://github.com/flamecodezz/llm-action/blob/main
大模型剪枝技术原理:SparseGPT、Wandahttps://github.com/flamecodezz/llm-action/blob/main
大模型剪枝技术原理:总结https://github.com/flamecodezz/llm-action/blob/main
https://github.com/flamecodezz/llm-action#llm知识蒸馏
大模型知识蒸馏概述https://www.zhihu.com/question/625415893/answer/3243565375
https://github.com/flamecodezz/llm-action#低秩分解
https://github.com/flamecodezz/llm-action#llm测评
C-Evalhttps://github.com/liguodongiot/ceval
CMMLUhttps://github.com/liguodongiot/CMMLU
IFEval: Instruction Following Evalhttps://github.com/google-research/google-research/tree/master/instruction_following_eval
Paperhttps://arxiv.org/abs/2311.07911
SuperCLUEhttps://github.com/CLUEbenchmark/SuperCLUE
AGIEvalhttps://github.com/ruixiangcui/AGIEval/
OpenCompasshttps://github.com/open-compass/opencompass/blob/main/README_zh-CN.md
https://github.com/flamecodezz/llm-action#llm数据工程
https://github.com/flamecodezz/llm-action#预训练语料处理技术
https://github.com/flamecodezz/llm-action/blob/main/pic/llm/train/pretrain/llm-pretrain-pipeline-v2.png
https://github.com/flamecodezz/llm-action#llm微调高效数据筛选技术
LLM微调高效数据筛选技术原理-DEITAhttps://github.com/flamecodezz/llm-action/blob/main
LLM微调高效数据筛选技术原理-MoDShttps://github.com/flamecodezz/llm-action/blob/main
LLM微调高效数据筛选技术原理-IFDhttps://github.com/flamecodezz/llm-action/blob/main
LLM微调高效数据筛选技术原理-CaRhttps://github.com/flamecodezz/llm-action/blob/main
LESS:仅选择5%有影响力的数据优于全量数据集进行目标指令微调https://zhuanlan.zhihu.com/p/686007325
LESS 实践:用少量的数据进行目标指令微调https://zhuanlan.zhihu.com/p/686687923
https://github.com/flamecodezz/llm-action#提示工程
LLM算法架构https://github.com/liguodongiot/llm-action/tree/main/docs/llm-base/ai-algo
https://github.com/flamecodezz/llm-action#llm算法架构
https://github.com/flamecodezz/llm-action/blob/main/pic/llm/model/llm-famliy.jpg
大模型算法演进https://zhuanlan.zhihu.com/p/600016134
https://github.com/flamecodezz/llm-action/blob/main/pic/llm/model/llm-timeline-v2.png
百川智能开源大模型baichuan-7B技术剖析https://www.zhihu.com/question/606757218/answer/3075464500
百川智能开源大模型baichuan-13B技术剖析https://www.zhihu.com/question/611507751/answer/3114988669
LLaMA3 技术剖析https://www.zhihu.com/question/653374932/answer/3470909634
https://github.com/flamecodezz/llm-action#llm应用开发
云原生向量数据库Milvus(一)-简述、系统架构及应用场景https://zhuanlan.zhihu.com/p/476025527
云原生向量数据库Milvus(二)-数据与索引的处理流程、索引类型及Schemahttps://zhuanlan.zhihu.com/p/477231485
关于大模型驱动的AI智能体Agent的一些思考https://zhuanlan.zhihu.com/p/651921120
LLM国产化适配https://github.com/liguodongiot/llm-action/tree/main/docs/llm_localization
https://github.com/flamecodezz/llm-action#llm国产化适配
大模型国产化适配1-华为昇腾AI全栈软硬件平台总结https://zhuanlan.zhihu.com/p/637918406
大模型国产化适配2-基于昇腾910使用ChatGLM-6B进行模型推理https://zhuanlan.zhihu.com/p/650730807
大模型国产化适配3-基于昇腾910使用ChatGLM-6B进行模型训练https://zhuanlan.zhihu.com/p/651324599
大模型国产化适配4-基于昇腾910使用LLaMA-13B进行多机多卡训练https://zhuanlan.zhihu.com/p/655902796
大模型国产化适配5-百度飞浆PaddleNLP大语言模型工具链总结https://zhuanlan.zhihu.com/p/665807431
大模型国产化适配6-基于昇腾910B快速验证ChatGLM3-6B/BaiChuan2-7B模型推理https://zhuanlan.zhihu.com/p/677799157
大模型国产化适配7-华为昇腾LLM落地可选解决方案(MindFormers、ModelLink、MindIE)https://zhuanlan.zhihu.com/p/692377206
MindIE 1.0.RC1 发布,华为昇腾终于推出了针对LLM的完整部署方案,结束小米加步枪时代https://www.zhihu.com/question/654472145/answer/3482521709
大模型国产化适配8-基于昇腾MindIE推理工具部署Qwen-72B实战(推理引擎、推理服务化)https://juejin.cn/post/7365879319598727180
大模型国产化适配9-LLM推理框架MindIE-Service性能基准测试https://zhuanlan.zhihu.com/p/704649189
大模型国产化适配10-快速迁移大模型到昇腾910B保姆级教程(Pytorch版)https://juejin.cn/post/7375351908896866323
大模型国产化适配11-LLM训练性能基准测试(昇腾910B3)https://juejin.cn/post/7380995631790964772
国产知名AI芯片厂商产品大揭秘-昇腾、海光、天数智芯...https://f46522gm22.feishu.cn/docx/PfWfdMKo8oXYN6xi7uycuhgFnKg
国内AI芯片厂商的计算平台大揭秘-昇腾、海光、天数智芯...https://f46522gm22.feishu.cn/docx/XnhcdXVDholUBpxYoMccS11Mnfc
【LLM国产化】量化技术在MindIE推理框架中的应用https://juejin.cn/post/7416723051377377316
⬆ 一键返回目录https://github.com/flamecodezz/llm-action#%E7%9B%AE%E5%BD%95
AI编译器https://github.com/liguodongiot/llm-action/tree/main/ai-compiler
https://github.com/flamecodezz/llm-action#ai编译器
AI编译器技术剖析(一)-概述https://zhuanlan.zhihu.com/p/669347560
AI编译器技术剖析(二)-传统编译器https://zhuanlan.zhihu.com/p/671477784
AI编译器技术剖析(三)-树模型编译工具 Treelite 详解https://zhuanlan.zhihu.com/p/676723324
AI编译器技术剖析(四)-编译器前端https://github.com/flamecodezz/llm-action/blob/main
AI编译器技术剖析(五)-编译器后端https://github.com/flamecodezz/llm-action/blob/main
AI编译器技术剖析(六)-主流编译框架https://github.com/flamecodezz/llm-action/blob/main
AI编译器技术剖析(七)-深度学习模型编译优化https://github.com/flamecodezz/llm-action/blob/main
lleaves:使用 LLVM 编译梯度提升决策树将预测速度提升10+倍https://zhuanlan.zhihu.com/p/672584013
https://github.com/flamecodezz/llm-action#ai基础设施
AI 集群基础设施 NVMe SSD 详解https://zhuanlan.zhihu.com/p/672098336
AI 集群基础设施 InfiniBand 详解https://zhuanlan.zhihu.com/p/673903240
大模型训练基础设施:算力篇https://github.com/flamecodezz/llm-action/blob/main
https://github.com/flamecodezz/llm-action#ai加速卡
AI芯片技术原理剖析(一):国内外AI芯片概述https://zhuanlan.zhihu.com/p/667686665
https://github.com/flamecodezz/llm-action#ai集群
AI集群网络通信https://github.com/liguodongiot/llm-action/tree/main/docs/llm-base/network-communication
https://github.com/flamecodezz/llm-action#ai集群网络通信
https://github.com/flamecodezz/llm-action#llmops
在 Kubernetes 上部署机器学习模型的指南https://zhuanlan.zhihu.com/p/676389726
使用 Kubernetes 部署机器学习模型的优势https://juejin.cn/post/7320513026188099619
https://github.com/flamecodezz/llm-action#llm生态相关技术
大模型词表扩充必备工具SentencePiecehttps://zhuanlan.zhihu.com/p/630696264
大模型实践总结https://www.zhihu.com/question/601594836/answer/3032763174
ChatGLM 和 ChatGPT 的技术区别在哪里?https://www.zhihu.com/question/604393963/answer/3061358152
现在为什么那么多人以清华大学的ChatGLM-6B为基座进行试验?https://www.zhihu.com/question/602504880/answer/3041965998
为什么很多新发布的大模型默认使用BF16而不是FP16?https://www.zhihu.com/question/616600181/answer/3195333332
大模型训练时ZeRO-2、ZeRO-3能否和Pipeline并行相结合?https://www.zhihu.com/question/652836990/answer/3468210626
一文详解模型权重存储新格式 Safetensorshttps://juejin.cn/post/7386360803039838235
一文搞懂大模型文件存储格式新宠GGUFhttps://juejin.cn/post/7408858126042726435
LLM面试题https://github.com/liguodongiot/llm-action/blob/main/llm-interview/README.md
https://github.com/flamecodezz/llm-action#llm面试题
大模型基础常见面试题https://github.com/liguodongiot/llm-action/blob/main/llm-interview/base.md
大模型算法常见面试题https://github.com/liguodongiot/llm-action/blob/main/llm-interview/llm-algo.md
大模型训练常见面试题https://github.com/liguodongiot/llm-action/blob/main/llm-interview/llm-train.md
大模型微调常见面试题https://github.com/liguodongiot/llm-action/blob/main/llm-interview/llm-ft.md
大模型评估常见面试题https://github.com/liguodongiot/llm-action/blob/main/llm-interview/llm-eval.md
大模型压缩常见面试题https://github.com/liguodongiot/llm-action/blob/main/llm-interview/llm-compress.md
大模型推理常见面试题https://github.com/liguodongiot/llm-action/blob/main/llm-interview/llm-inference.md
大模型应用常见面试题https://github.com/liguodongiot/llm-action/blob/main/llm-interview/llm-app.md
大模型综合性面试题https://github.com/liguodongiot/llm-action/blob/main/llm-interview/comprehensive.md
⬆ 一键返回目录https://github.com/flamecodezz/llm-action#%E7%9B%AE%E5%BD%95
https://github.com/flamecodezz/llm-action#服务器基础环境安装及常用工具
英伟达A800加速卡常见软件包安装命令https://github.com/liguodongiot/llm-action/blob/main/docs/llm-base/a800-env-install.md
英伟达H800加速卡常见软件包安装命令https://github.com/liguodongiot/llm-action/blob/main/docs/llm-base/h800-env-install.md
昇腾910加速卡常见软件包安装命令https://github.com/liguodongiot/llm-action/blob/main/llm_localization/ascend910-env-install.md
Linux 常见命令大全https://juejin.cn/post/6992742028605915150
Conda 常用命令大全https://juejin.cn/post/7089093437223338015
Poetry 常用命令大全https://juejin.cn/post/6999405667261874183
Docker 常用命令大全https://juejin.cn/post/7016238524286861325
Docker Dockerfile 指令大全https://juejin.cn/post/7016595442062327844
Kubernetes 常用命令大全https://juejin.cn/post/7031201391553019911
集群环境 GPU 管理和监控工具 DCGM 常用命令大全https://github.com/liguodongiot/llm-action/blob/main/docs/llm-base/dcgmi.md
https://github.com/flamecodezz/llm-action#llm学习交流群
https://github.com/liguodongiot/llm-action/blob/main/pic/wx.jpg
https://github.com/flamecodezz/llm-action#微信公众号
https://github.com/liguodongiot/llm-action/blob/main/pic/wx-gzh.png
⬆ 一键返回目录https://github.com/flamecodezz/llm-action#%E7%9B%AE%E5%BD%95
https://github.com/flamecodezz/llm-action#star-history
https://star-history.com/#liguodongiot/llm-action&Date
https://github.com/flamecodezz/llm-action#ai工程化课程推荐
llm-resourcehttps://github.com/liguodongiot/llm-resource
ai-systemhttps://github.com/liguodongiot/ai-system
点击https://mp.weixin.qq.com/s?__biz=MzU3Mzg5ODgxMg==&mid=2247488019&idx=1&sn=90ca4657a643431f21df44fa199e695f&chksm=fd3bfb40ca4c72565fdb91717c8a27739bcb71d10fae89c7994f141d375d21059be19e702164&token=701439972&lang=zh_CN#rd
www.zhihu.com/column/c_1456193767213043713https://www.zhihu.com/column/c_1456193767213043713
Readmehttps://github.com/flamecodezz/llm-action#readme-ov-file
Licensehttps://github.com/flamecodezz/llm-action#License-1-ov-file
Activityhttps://github.com/flamecodezz/llm-action/activity
0 forkshttps://github.com/flamecodezz/llm-action/forks
Report repositoryhttps://github.com/contact/report-content?content_url=https%3A%2F%2Fgithub.com%2Fflamecodezz%2Fllm-action&report=flamecodezz+%28user%29
https://github.com
Termshttps://docs.github.com/site-policy/github-terms/github-terms-of-service
Privacyhttps://docs.github.com/site-policy/privacy-policies/github-privacy-statement
Securityhttps://github.com/security
Statushttps://www.githubstatus.com/
Communityhttps://github.community/
Docshttps://docs.github.com/
Contacthttps://support.github.com?tags=dotcom-footer

Response headers:
Servergithub.com
Content-Typetext/html; charset=utf-8
Cache-Controlmax-age=0, private, must-revalidate

Viewport: width=device-width


URLs of crawlers that visited me.