From ce8a3eae5be301d1e3f536ede32a2b0c65f7c51f Mon Sep 17 00:00:00 2001
From: Sze-qq <68757353+Sze-qq@users.noreply.github.com>
Date: Mon, 4 Apr 2022 13:47:43 +0800
Subject: [PATCH] update GPT-2 experiment result (#666)
---
README-zh-Hans.md | 5 +++--
README.md | 5 +++--
2 files changed, 6 insertions(+), 4 deletions(-)
diff --git a/README-zh-Hans.md b/README-zh-Hans.md
index 7ccdeaa19..5e15cd5af 100644
--- a/README-zh-Hans.md
+++ b/README-zh-Hans.md
@@ -86,9 +86,10 @@ Colossal-AI为您提供了一系列并行训练组件。我们的目标是让您
- 降低11倍GPU显存占用，或超线性扩展（张量并行）
-
+GPT-2.png)
-- 能训练接近11倍大小的模型（ZeRO）
+- 用相同的硬件条件训练24倍大的模型
+- 超3倍的吞吐量
### BERT
diff --git a/README.md b/README.md
index 400116d1d..1378d8f24 100644
--- a/README.md
+++ b/README.md
@@ -87,9 +87,10 @@ distributed training in a few lines.
- 11x lower GPU memory consumption, and superlinear scaling efficiency with Tensor Parallelism
-
+
-- 10.7x larger model size on the same hardware
+- 24x larger model size on the same hardware
+- over 3x acceleration
### BERT