Update README.md

README.md

@@ -5,12 +5,18 @@ language:
 - en
 base_model:
 - meta-llama/Llama-3.1-8B
+tags:
+- biology
+- medical
+- Llama
 ---
 
 ![image/png](https://cdn-uploads.huggingface.co/production/uploads/646704281dd5854d4de2cdda/vBbfKhWIBJVIjqSv6yVDk.png)
 
 # KoBioMed-Llama-3.1-8B
 
+_**The model weights are scheduled to be uploaded on March 17, 2025.**_
+
 ## Instroduction
 We introduce KoBioMed-Llama-3.1-8B, a bilingual (English and Korean) generative model specialized in the BioMedical domain, developed by ezCaretech. This model has been continual pre-trained (CPT) on a dataset from PubMed abstracts and their translated Korean counterparts, undergoing extensive preprocessing that includes cleansing, de-duplication, and quality filtering.
 
@@ -24,12 +30,16 @@ This repository contains an 8 Billion generative language model with the followi
 - Vocab Size: 12,800
 - License: llama3.1
 
-**This model was developed with support from the Korea Artificial Intelligence Industry Cluster Agency (AICA).**
+
 
 
 
 ### Notice!
-The model is currently in post-training (Instruction Tuning, DPO) and is scheduled to be released within March 2025.
+- **This is a pre-trained model. It will be a great starting point for post-training, such as instruction tuning.**
+
+- **This model was developed with support from the Korea Artificial Intelligence Industry Cluster Agency (AICA).**
+
+- **The model is currently in post-training (Instruction Tuning, DPO) and is scheduled to be released within March 2025.**
 
 
 ## Evaluation
@@ -48,23 +58,23 @@ We evaluated the KoBioMed-Llama-3.1-8B using various Korean and English biomedic
         <th>MedMCQA</th>
         <th>MMLU</th>
         <th>PubMedQA</th>
-        <th>Average</th>
+        <th>Mean</th>
     </tr>
     <tr>
-        <td>KoBioMed-Llama-3.1-8B</td>
-        <td align="center"><strong>0.0</strong></td>
-        <td align="center"><strong>0.0</strong></td>
-        <td align="center"><strong>0.0</strong></td>
-        <td align="center"><strong>0.0</strong></td>
-        <td align="center"><strong>0.0</strong></td>
-        <td align="center"><strong>0.0</strong></td>
+        <td>KoBioMed-Llama-3.1-8B (v0.5)</td>
+        <td align="center"><strong>0.3930</strong></td>
+        <td align="center"><strong>0.5641</strong></td>
+        <td align="center"><u>0.5326</u></td>
+        <td align="center">0.6771</td>
+        <td align="center"><strong>0.7940</strong></td>
+        <td align="center"><strong>0.5922</strong></td>
     </tr>
     <tr>
         <td>Llama-3.1-8B</td>
-        <td align="center">0.3620</td>
+        <td align="center"><u>0.3620</u></td>
         <td align="center">0.5105</td>
-        <td align="center">0.5635</td>
-        <td align="center">0.7159</td>
+        <td align="center"><strong>0.5635</strong></td>
+        <td align="center"><strong>0.7159</strong></td>
         <td align="center">0.7600</td>
         <td align="center">0.5824</td>
     </tr>
@@ -89,10 +99,10 @@ We evaluated the KoBioMed-Llama-3.1-8B using various Korean and English biomedic
     <tr>
         <td>SOLAR-10.7B-v1.0</td>
         <td align="center">0.3200</td>
-        <td align="center">0.5146</td>
+        <td align="center"><u>0.5146</u></td>
         <td align="center">0.5075</td>
-        <td align="center">0.7050</td>
-        <td align="center">0.7760</td>
+        <td align="center"><u>0.7050</u></td>
+        <td align="center"><u>0.7760</u></td>
         <td align="center">0.5646</td>
     </tr>
 </table>