Update README.md

README.md

@@ -19,7 +19,7 @@ Datasets have been translated with google-translate-api for reward training:
 - [Dahoas/synthetic-instruct-gptj-pairwise](https://huggingface.co/datasets/Dahoas/synthetic-instruct-gptj-pairwise)
 - [openai/webgpt_comparisons](https://huggingface.co/datasets/openai/webgpt_comparisons)
 
-Fisrtly download custom model localy. You can do it manualy.
+Firstly download custom model localy. You can do it manualy.
 
 OR:
 - git lfs install;