From b43dc1c5ca527ddc643b2deff7263a19f58e4613 Mon Sep 17 00:00:00 2001 From: mst272 Date: Thu, 18 Jul 2024 19:40:59 +0800 Subject: [PATCH] update transformer --- README.md | 1 + llm_tricks/transformer/README.md | 3 +++ 2 files changed, 4 insertions(+) create mode 100644 llm_tricks/transformer/README.md diff --git a/README.md b/README.md index 63d65f2..2feda7f 100644 --- a/README.md +++ b/README.md @@ -26,6 +26,7 @@ Tips: 图片完全由AI生成 - [致谢](#-致谢) ## 📖 Latest News +- [2024-07-19] 🤓RLHF 强化学习框架新增CPO,SimPO,以及二者融合CPO-SimPO - [2024-07-16] 🤓RLHF 强化学习框架更新完成,支持deepspeed单卡/多卡 进行强化学习lora、qlora等训练,详细可见[RLHF](./rlhf/README.md) - [2024-06-10] 🚀增加一步一步实现Transformer技术发文(包括代码等从零介绍),可见 [技术发文](#技术发文) - [2024-06-9] 🚀支持DPO训练,分为单轮对话DPO(自己构建,方便魔改)和多轮对话DPO(简洁实现),支持deepspeed的lora和qlora,具体介绍可见 [DPO使用说明](./train_args/dpo/README.md) diff --git a/llm_tricks/transformer/README.md b/llm_tricks/transformer/README.md new file mode 100644 index 0000000..5a23c2a --- /dev/null +++ b/llm_tricks/transformer/README.md @@ -0,0 +1,3 @@ +# Transformer代码详解:从头开始实现 + +全部代码:https://github.com/mst272/transformer-pytorch \ No newline at end of file