Update README.md

README.md

@@ -15,7 +15,10 @@ Based on [LLaMA 7B](https://huggingface.co/decapoda-research/llama-7b-hf).
 * Merged version: [link](https://huggingface.co/IlyaGusev/llama_7b_ru_turbo_alpaca_lora_merged).
 * llama.cpp version: [link](https://huggingface.co/IlyaGusev/llama_7b_ru_turbo_alpaca_lora_llamacpp).
 
-Warning! The model was trained with a target capped at 256 tokens. We will update it once a version with 512 tokens is ready.
+April 2, 2023: A new model version was uploaded, trained with outputs capped at 512 tokens instead of 256. To use the old version, specify
+```
+revision="dd4b2c63d6639fd9490cd1d9b0d6d8b58e708d93"
+```
 
 Colab: [link](https://colab.research.google.com/drive/1JLoHOjDJQIa8SDqsEXrGHj4Z4aTnaajN)