From 61b8b42b3c04ab9783f276913864c9c499ac98fa Mon Sep 17 00:00:00 2001 From: "hanluzhi0429@gmail.com" Date: Mon, 26 May 2025 22:35:30 +0800 Subject: [PATCH] [25.05.26] add documents->tokenizer --- README.md | 3 +++ 1 file changed, 3 insertions(+) diff --git a/README.md b/README.md index 509d4dd..1b5c57b 100755 --- a/README.md +++ b/README.md @@ -4,6 +4,9 @@ LLM-Dolly is a custom LLM built from scratch ## 🗓️ 更新日志 +#### 2025-05-26 +- 📝 增加LLM文档，涵盖LLM全周期技术点。 + #### 2025-05-22 - 📝 添加支持MoE模型，训练资源占用不稳定，测试模型Experts=8, experts_per_tok=4,GPU显存从60%-94%跳动,单个GPU利用率0%-100%跳动。当调小batch_size后，显存占用稳定在90%，单个GPU利用率稳定在90%以上，偶现30%的利用率。