本仓库定位为 AI论文复现 / 从零实现 Transformer。 代码遵循原论文的模块划分，包含位置编码、多头注意力、前馈网络、编码器‑解码器等全部组件，并附带详细的中文拆解文档与英文注释，方便学习与二次开发。
☆41May 3, 2025Updated last year
Alternatives and similar repositories for Transformer-from-Scratch
Users that are interested in Transformer-from-Scratch are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- 本项目从零开始构建并优化了一个千万参数级别的大规模预训练语言模型，涵盖预训练、有监督微调（SFT）和R1推理蒸馏三个阶段。项目采用自定义Transformer架构（包括RMSNorm、分组注意力、多Query 机制、SwiGLU激活和RoPE位置编码），实现高效的长文本处理和…☆23Mar 10, 2025Updated last year
- BUAA Compiler Course Project 2023 by Toby Shi.☆13Aug 20, 2024Updated 2 years ago
- 深澜校园网认证☆13Jun 13, 2022Updated 4 years ago
- 操作系统实验 进程管理☆13May 24, 2017Updated 9 years ago
- github Actions王者荣耀 王者营地自动签到每周浏览点赞任务☆14Mar 25, 2021Updated 5 years ago
- Deploy to Railway using AI coding agents - Free Credits Offer • AdUse Claude Code, Codex, OpenCode, and more. Autonomous software development now has the infrastructure to match with Railway.
- The objective of this project is to demonstrate how to fine-tune deepseek-r1-distill-llama-8b.☆17Feb 19, 2025Updated last year
- 2022秋季学期-北航计院-编译原理实验课设☆12Jun 25, 2023Updated 3 years ago
- 影速（GoComicMosaic的独立子项目）是一个用于测试影视采集接口速度和稳定性的工具。通过批量测试多个采集接口的响应时间、成功率和结果数量，进行综合打分排序。☆31Jun 17, 2025Updated last year
- [Main EMNLP'25] LLMs do Multi-Label Classification Differently☆20Feb 28, 2026Updated 6 months ago
- 这是大连理工大学电计2203班的「知识小料」系列，一般在每个教学周更新一期。☆16Jan 30, 2026Updated 7 months ago
- ☆10Apr 5, 2023Updated 3 years ago
- 测试脚本和框架代码分离的轻量级接口自动化框架： 1、能够管理系统公共的数据，系统发生变化后方便切换 2、能够实现接口层级的复用，提高接口自动化脚本的可维护性 3、能够实现接口之间的关联，也就是需要结果提取器 4、能够操作数据库处理数据（预制数据、清理垃圾数据） 5、能够实现…☆11Mar 25, 2021Updated 5 years ago
- 一款用python写的接口测试框架，可以对单接口 和多接口逻辑进行全方位的测试。框架分为三部分：1.执行前 2.执行中 3.执行后☆10May 24, 2017Updated 9 years ago
- AI empowers a general-purpose prompt assistant, systematically generating and selecting prompts on AI generation platforms such as Midjou…☆12Jan 8, 2026Updated 8 months ago
- GPU virtual machines on DigitalOcean Gradient AI • AdGet to production fast with high-performance AMD and NVIDIA GPUs you can spin up in seconds. The definition of operational simplicity.
- 一个基于 bilibili 用户行为数据和深度学习的个性化视频推荐系统。☆27Mar 12, 2025Updated last year
- Markdown转小红书卡片☆10Apr 10, 2025Updated last year
- MG400/M1Pro; Python package based on TCP/IP protocol☆30Jul 30, 2026Updated last month
- Lite-MCP-Client是一个基于命令行的轻量级MCP客户端工具☆16Apr 1, 2025Updated last year
- ☆10Feb 27, 2025Updated last year
- ☆28Sep 10, 2026Updated 2 weeks ago
- 微信小程序应用开发赛（后台）☆10Oct 24, 2020Updated 5 years ago
- [NeurIPS 2025] Reasoning MLLM, Share-GRPO, advantage vanishing, sparse reward☆38Sep 19, 2025Updated last year
- Android MCP Server implementation☆15Jul 11, 2026Updated 2 months ago
- Managed Database hosting by DigitalOcean • AdPostgreSQL, MySQL, MongoDB, Kafka, Valkey, and OpenSearch available. Automatically scale up storage and focus on building your apps.
- ☆10Jun 29, 2022Updated 4 years ago
- HttpRunner V1.x 中文使用文档☆11Dec 6, 2021Updated 4 years ago
- 极客时间Java进阶训练营学习笔记☆15Jan 31, 2021Updated 5 years ago
- 仿vue-element-admin 项目 预览地址：☆10Aug 4, 2020Updated 6 years ago
- 王者荣耀全自动获取金币☆27Apr 25, 2025Updated last year
- Learning to Retrieve by Trying - Source code for Grounding by Trying: LLMs with Reinforcement Learning-Enhanced Retrieval☆52Oct 31, 2024Updated last year
- An example of External Game Hacking (CSGO) using CTHackFramework.☆44May 31, 2019Updated 7 years ago
- 📸 Crawell – 网页图片/正文一键提取、Markdown 转换与批量下载的浏览器扩展，本地化，免费 Crawell browser extension for one-click image & article extraction, Markdown con…☆16Jul 31, 2025Updated last year
- ☆12Dec 16, 2020Updated 5 years ago
- Simple, predictable pricing with DigitalOcean hosting • AdAlways know what you'll pay with monthly caps and flat pricing. Enterprise-grade infrastructure trusted by 600k+ customers.
- python rest framework + vue + elementUI 前后端分离接口自动化平台☆14Jan 4, 2021Updated 5 years ago
- Django Auto Test☆11Aug 25, 2019Updated 7 years ago
- ☆15Sep 11, 2025Updated last year
- 用于提取推特页面中包含的媒体文件 (支持文本, 图片, 视频, 动图) 的脚本☆46Sep 29, 2022Updated 3 years ago
- edge-tts webui☆13Jul 22, 2023Updated 3 years ago
- ☆12Aug 9, 2019Updated 7 years ago
- zenfeed Official lightweight WEB client☆19Feb 4, 2026Updated 7 months ago