Update README.md

README.md

@@ -9,13 +9,15 @@ inference: false
 
 # LLaMA 7B trained on the ru_turbo_alpaca, Russian instructions dataset
 
+Based on [LLaMA 7B](https://huggingface.co/decapoda-research/llama-7b-hf).
+
 An adapter only version. Merged version: [link](https://huggingface.co/IlyaGusev/llama_7b_ru_turbo_alpaca_lora_merged).
 
 Warning! The model was trained with a target capped at 256 tokens. We will update it once a version with 512 tokens is ready.
 
 Colab: [link](https://colab.research.google.com/drive/1JLoHOjDJQIa8SDqsEXrGHj4Z4aTnaajN)
 
-Training code: [rulm/self_instruct](https://github.com/IlyaGusev/rulm/tree/master/self_instruct)
+Training code: [link](https://github.com/IlyaGusev/rulm/tree/master/self_instruct)
 
 ```python
 from peft import PeftModel, PeftConfig