Update README.md

README.md

@@ -119,7 +119,7 @@ print(scores) # [0.0020704232156276703, 0.9999990463256836] first document is no
 
 # Training
 
-We used [LLaMA-Factory](https://github.com/hiyouga/LLaMA-Factory) to fine-tune Mistral to create FollowIR-7B, with the following training script:
+We used [LLaMA-Factory](https://github.com/hiyouga/LLaMA-Factory) to fine-tune Mistral to create FollowIR-7B, after transforming it to fit their format (input of "query" + "instruction" inside the template, output is the label, and instruction as the beginning of the template) with the following training script:
 ```bash
 #!/bin/bash
 accelerate launch src/train_bash.py \