Update README.md

README.md

@@ -1,9 +1,14 @@
 ---
 library_name: peft
+language:
+- ru
 ---
-## Training procedure
+Use in the same way as IlyaGusev/saiga2_7b_lora.
+WARNING! Load tokenizer as AutoTokenizer.from_pretrained(model_path, use_fast=True)
 
-### Framework versions
+Up to 60% faster generation and 35% training with HF because of different tokenizer.
 
+## Model description
 
-- PEFT 0.5.0
+Instruction version (Saiga datasets) of Russian adaptation of LLaMa-2-7B by replacing the tokenizer.
+Paper: Tikhomirov M.M., Chernyshev D.I., Impact of Tokenization on LLaMa Russian Adaptation (will be soon)