🧠 两小时从 0 训出你的 64M 大模型。Train your own 64M LLM from scratch in 2 hours.
[feat] update readme
jingyaogongea2abb5fb317
diff --git a/README.md b/README.mdindex e35e129..ee3b8a1 100644--- a/README.md+++ b/README.md@@ -388,11 +388,13 @@ torchrun --nproc_per_node N train_xxx.py<details style="color:rgb(128,128,128)"><summary>注:其它须知</summary>+<del>单机N卡启动训练 (DeepSpeed)```bashdeepspeed --master_port 29500 --num_gpus=N train_xxx.py```+</del>可根据需要开启wandb记录训练过程(需可直连)diff --git a/README_en.md b/README_en.mdindex 366af4a..4c2759c 100644--- a/README_en.md+++ b/README_en.md@@ -381,11 +381,13 @@ torchrun --nproc_per_node N train_xxx.py<details style="color:rgb(128,128,128)"><summary>Note: Other Notes</summary>+<del>Single machine N GPU training (DeepSpeed)```bashdeepspeed --master_port 29500 --num_gpus=N train_xxx.py```+</del>You can optionally enable wandb to record the training process (requires direct internet connection)