From e4d1bf72b34b2266b18c48624410c327f634180a Mon Sep 17 00:00:00 2001 From: "hanluzhi0429@gmail.com" Date: Thu, 19 Jun 2025 18:25:21 +0800 Subject: [PATCH] [2025.06.19] pretoken corpus and add cold start --- README.md | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/README.md b/README.md index 68d609e..ba95ac4 100755 --- a/README.md +++ b/README.md @@ -4,12 +4,15 @@ LLM-Dolly is a custom LLM built from scratch 🗓️ 更新日志 -#### 2025-06-05 -- 📝 上传tokenizer的训练权重(1.56字符/token),调整tokenizer的格式和训练方式,和qwen2的tokenizer风格保持一致. +#### 2025-06-19 +- 📝 pretoken corpus and update cold start code.
查看更多 +#### 2025-06-05 +- 📝 上传tokenizer的训练权重(1.56字符/token),调整tokenizer的格式和训练方式,和qwen2的tokenizer风格保持一致. + #### 2025-05-26 - 📝 增加LLM文档,涵盖LLM全周期技术点。