Update README.md

README.md

@@ -5,7 +5,7 @@ datasets:
 language:
 - en
 base_model:
-- meta-llama/Llama-3.1-8B-Instruct
+- trendmicro-ailab/Llama-Primus-Merged
 pipeline_tag: text-generation
 library_name: transformers
 tags:
@@ -16,11 +16,11 @@ extra_gated_fields:
   Country: country
   I want to use this model for:
     type: select
-    options: 
-      - Research
-      - Commercial
-      - label: Other
-        value: other
+    options:
+    - Research
+    - Commercial
+    - label: Other
+      value: other
   Job title:
     type: select
     options:
@@ -38,7 +38,7 @@ extra_gated_fields:
 
 <img src="https://i.imgur.com/PtqeTZw.png" alt="Primus Overview" width="60%">
 
->TL;DR: Llama-Primus-Reasoning is a reasoning model distilled from the reasoning steps with reflection data generated by o1-preview on cybersecurity tasks (_Primus-Reasoning_). It demonstrates a 🚀**10%** improvement in security certification (CISSP).
+>TL;DR: Llama-Primus-Reasoning is a reasoning model distilled from the reasoning steps with reflection data generated by o1-preview on cybersecurity tasks (_Primus-Reasoning_), with Llama-Primus-Merged as the base model for distillation. It demonstrates a 🚀**10%** improvement in security certification (CISSP).
 
 **🔥 For more details, please refer to the paper: [[📄Paper]](https://arxiv.org/abs/2502.11191).**