Update README.md

README.md

@@ -49,17 +49,17 @@ SFT LoRA обучение было выполнено на **двух NVIDIA A10
 - **Language(s) (NLP):** [RU/EN]
 - **Finetuned from model:** [Qwen2.5]  
 
-**Distributed training:**
-- DeepSpeed (Stage 3)
-- HuggingFace Accelerator
-
-**Fusion:**
+**Distributed training:**  
+- DeepSpeed (Stage 3)  
+- HuggingFace Accelerator  
+  
+**Fusion:**  
 - Flash Attention 2  
 - Fused AdamW  
-- Liger Kernel (swiglu, fused linear xentropy)
-- 
-**GPU hours**: ~384 of NVIDIA A100
-
+- Liger Kernel (swiglu, fused linear xentropy)  
+  
+**GPU hours**: ~384 of NVIDIA A100  
+  
 ### Training configuration / Конфигурация обучения
 **The model was trained using MyLLM framework:**  
 --== [MyLLM](https://github.com/Raumberg/myllm) ==--