Update README.md

README.md

@@ -10,19 +10,7 @@ license: apache-2.0
   <a href="https://infix-ai.com/research/infir2/">🌐 Project Website</a> &nbsp;
 </p>
 
-We performed supervised fine-tuning on the **InfiR2-7B-base-FP8** with FP8 format in two stages using the InfiAlign-SFT-72k and InfiAlign-SFT-165k datasets, with hyperparameters shown in below.
-
-<div align="center">
-
-| Parameter | Value |
-| :---: | :---: |
-| **Batch Size** | 64 |
-| **Learning Rate** | 1e-5 |
-| **Minimum Learning Rate** | 1e-6 |
-| **Weight Decay** | 0.05 |
-| **Context Length** | 32k |
-
-</div>
+We performed supervised fine-tuning on the **InfiR2-7B-base-FP8** with FP8 format in two stages using the InfiAlign-SFT-72k and InfiAlign-SFT-165k datasets.
 
 The resulting model is the **InfiR2-7B-Instruct-FP8**.
 
@@ -35,6 +23,19 @@ The resulting model is the **InfiR2-7B-Instruct-FP8**.
 - Stable and Reproducible Performance
 - Efficient and Low memory Training 
 
+**Hyperparameters**:
+
+<div align="center">
+
+| Parameter | Value |
+| :---: | :---: |
+| **Batch Size** | 64 |
+| **Learning Rate** | 1e-5 |
+| **Minimum Learning Rate** | 1e-6 |
+| **Weight Decay** | 0.05 |
+| **Context Length** | 32k |
+
+</div>
 
 
 ## 🚀 InfiR2 Model Series