[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-74114":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":9,"language":10,"languages":9,"totalLinesOfCode":9,"stars":11,"forks":12,"watchers":13,"openIssues":14,"contributorsCount":15,"subscribersCount":15,"size":15,"stars1d":16,"stars7d":17,"stars30d":18,"stars90d":15,"forks30d":15,"starsTrendScore":17,"compositeScore":19,"rankGlobal":9,"rankLanguage":9,"license":20,"archived":21,"fork":21,"defaultBranch":22,"hasWiki":23,"hasPages":21,"topics":24,"createdAt":9,"pushedAt":9,"updatedAt":25,"readmeContent":26,"aiSummary":27,"trendingCount":15,"starSnapshotCount":15,"syncStatus":28,"lastSyncTime":29,"discoverSource":30},74114,"smol-course","huggingface\u002Fsmol-course","huggingface","A course on aligning smol models.",null,"Jupyter Notebook",6657,2282,44,38,0,3,9,15,78,"Apache License 2.0",false,"main",true,[],"2026-06-12 04:01:13","![smolcourse image](.\u002Fbanner.png)\n\n# a smol course\n\nThis is a practical course on aligning language models for your specific use case. It's a handy way to get started with aligning language models, because everything runs on most local machines. There are minimal GPU requirements and no paid services. The course is built around the [SmolLM3](https:\u002F\u002Fhuggingface.co\u002Fblog\u002Fsmollm3) and [SmolVLM2](https:\u002F\u002Fhuggingface.co\u002Fblog\u002Fsmolvlm2) models, but the skills you'll learn can be applied to larger models or other small LLMs\u002FVLMs as well.\n\n\u003Cdiv style=\"background: linear-gradient(to right, #e0f7fa, #e1bee7, orange); padding: 20px; border-radius: 5px; margin-bottom: 20px; color: purple;\">\n    \u003Ch2>smol course v2 is live!\u003C\u002Fh2>\n    \u003Cp>This course is open and peer reviewed. To get involved with the course \u003Cstrong>open a pull request\u003C\u002Fstrong> and submit your work for review. Here are the steps:\u003C\u002Fp>\n    \u003Col>\n        \u003Cli>Follow the \u003Ca href=\"https:\u002F\u002Fhuggingface.co\u002Fsmol-course\">Hugging Face Hub org\u003C\u002Fa>\u003C\u002Fli>\n        \u003Cli>Read the material, make changes, do the exercises, add your own examples.\u003C\u002Fli>\n        \u003Cli>Submit a model to the leaderboard\u003C\u002Fli>\n        \u003Cli>Climb the leaderboard\u003C\u002Fli>\n    \u003C\u002Fol>\n    \u003Cp>This should help you learn and to build a community-driven course that is always improving.\u003C\u002Fp>\n\u003C\u002Fdiv>\n\n\n\u003Ca href=\"http:\u002F\u002Fhf.co\u002Fjoin\u002Fdiscord\">\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDiscord-7289DA?&logo=discord&logoColor=white\"\u002F>\n\u003C\u002Fa>\n\n\n\n\n## Future of this course\n\nThis course will soon be re-released on Hugging Face Learn! Stay tuned for updates.\n\n## Course Outline\n\nThis course provides a practical, hands-on approach to working with small language models, from initial training through to production deployment.\n\n| # | Topic | Description | Released |\n| - | ----- | ----------- | -------- |\n| 1 | Instruction Tuning | Supervised fine-tuning, chat templates, instruction following | ✅ |\n| 2 | Evaluation | Benchmarks and custom domain evaluation | ✅ |\n| 3 | Preference Alignment | Aligning models to human preferences with algorithms like DPO. | ✅ |\n| 4 | Vision Language Models | Adapt and use multimodal models | ✅ |\n| 5 | Reinforcement Learning | Optimizing models with based on reinforcement policies. | October 2025 |\n| 6 | Synthetic Data | Generate synthetic datasets for custom domains | November 2025 |\n| 7 | Award Ceremony | Showcase projects and celebrate | December 2025 |\n\n\n## Why Small Language Models?\n\nWhile large language models have shown impressive capabilities, they often require significant computational resources and can be overkill for focused applications. Small language models offer several advantages for domain-specific applications:\n\n- **Efficiency**: Require significantly less computational resources to train and deploy\n- **Customization**: Easier to fine-tune and adapt to specific domains\n- **Control**: Better understanding and control of model behavior\n- **Cost**: Lower operational costs for training and inference\n- **Privacy**: Can be run locally without sending data to external APIs\n- **Green Technology**: Advocates efficient usage of resources with reduced carbon footprint\n- **Easier Academic Research Development**: Provides an easy starter for academic research with cutting-edge LLMs with less logistical constraints\n\n## Prerequisites\n\nBefore starting, ensure you have the following:\n- Basic understanding of machine learning and natural language processing.\n- Familiarity with Python, PyTorch, and the `transformers` library.\n- Access to a pre-trained language model and a labeled dataset.\n\n## v1 of the course\n\nThe first version of the course used GithHub markdown and Jupyter notebooks. You can find it in the [v1](.\u002Fv1) directory.\n\n","该项目是一个关于如何为特定应用场景调整小型语言模型的实用课程。它基于SmolLM3和SmolVLM2模型，提供了从初步训练到生产部署的手把手指导，涵盖了指令调优、评估、偏好对齐、视觉语言模型等主题，并计划未来加入强化学习与合成数据生成等内容。本课程几乎可以在任何本地机器上运行，对GPU要求低且无需付费服务，特别适合希望以较低成本快速入门语言模型调整的学习者或开发者。此外，该课程开放同行评审，鼓励社区参与贡献，有助于构建一个持续改进的学习资源。",2,"2026-06-11 03:48:52","high_star"]