Update README.md

README.md

@@ -16,7 +16,7 @@ T-pro-it-2.0 is a model built upon the Qwen 3 model family and incorporates both
 ### 📚 Dataset
 
 Instruction Pre-Training: 
-40B tokens of instruction data, with one-third focused on reasoning tasks.
+40B tokens of instruction data, with about 20% focused on reasoning tasks.
 
 Supervised Fine-Tuning (SFT): 
 ~500K high-quality and diverse instructions with balanced complexity. Reasoning tasks make up about 20% of the dataset.