🧠 两小时从 0 训出你的 64M 大模型。Train your own 64M LLM from scratch in 2 hours.
update readme
jingyaogong00d145c481c1
diff --git a/README.md b/README.mdindex 21f949b..c92ac0f 100644--- a/README.md+++ b/README.md@@ -130,7 +130,6 @@- generate方式重构,继承自GenerationMixin类。- 🔥支持llama.cpp、vllm、ollama等热门三方生态。- 规范代码和目录结构。-- 🔥更新:从0实现PPO、GRPO的训练代码。- 改动词表`<s></s>`->`<|im_start|><|im_end|>````text为兼容第三方推理框架llama.cpp、vllm,本次更新需付出一些可观代价。@@ -510,12 +509,12 @@ quality(当然也还不算high,提升数据质量无止尽)。----## Ⅷ 数据集下载+## Ⅷ MiniMind训练数据集> [!NOTE]> 2025-02-05后,开源MiniMind最终训练所用的所有数据集,因此无需再自行预处理大规模数据集,避免重复性的数据处理工作。-MiniMind训练数据集 ([ModelScope](https://www.modelscope.cn/datasets/gongjy/minimind_dataset/files) | [HuggingFace](https://huggingface.co/datasets/jingyaogong/minimind_dataset/tree/main))+MiniMind训练数据集下载地址: [HuggingFace](https://huggingface.co/datasets/jingyaogong/minimind_dataset/tree/main)> 无需全部clone,可单独下载所需的文件diff --git a/README_en.md b/README_en.mdindex 1412244..7550da2 100644--- a/README_en.md+++ b/README_en.md@@ -145,9 +145,7 @@ We hope this open-source project can help LLM beginners quickly get started!• 🔥 Support for popular third-party ecosystems like llama.cpp, vllm, and ollama.-• Standardized code and directory structure.--• 🔥 New: Added training code for PPO and GRPO from scratch.+• Standardized code and directory structure.• Updated vocabulary tokens: `<s></s>` → `<|im_start|><|im_end|>`.@@ -559,7 +557,7 @@ Big respect!----## Ⅷ Dataset Download+## Ⅷ MiniMind Dataset Download> [!NOTE]> After `2025-02-05`, MiniMind’s open-source datasets for final training are provided, so there is no need for@@ -567,7 +565,7 @@ Big respect!MiniMind Training Datasets are available for download from:-Dataset ([ModelScope](https://www.modelscope.cn/datasets/gongjy/minimind_dataset/files) | [HuggingFace](https://huggingface.co/datasets/jingyaogong/minimind_dataset/tree/main))+MiniMind Dataset ([HuggingFace](https://huggingface.co/datasets/jingyaogong/minimind_dataset/tree/main))> You don’t need to clone everything, just download the necessary files.