Update README.md

README.md

@@ -15,9 +15,14 @@ base_model:
 ModularStarEncoder-finetuned-18 is an encoder built on top of [ModularStarEncoder-1B Pre-trained](https://huggingface.co/andreagurioli1995/ModularStarEncoder) on [SynthCode2Code2NL](https://huggingface.co/datasets/andreagurioli1995/SynthCode2Code2NL-neardedup). 
 ModularStarEncoder fine-tuned-18 is an encoder for various retrieval tasks, enabling the end user to select the model size that meets their memory and computational constraints.
 We built ModularStarEncoder on top of [StarCoder-2](https://huggingface.co/bigcode/starcoder2-15b), reducing its size from 15B to 1B parameters in bfloat16.
+
 This version contains only the first 18 layers of ModularStarEncoder-finetuned, with the related projection head.
 We have released this version to enhance the model's usability by allowing users to download only the desired size. 
-The model is finetuned with [CLIP objective](https://github.com/mlfoundations/open_clip/blob/main/src/open_clip/loss.py)
+
+The model is finetuned with [CLIP objective](https://github.com/mlfoundations/open_clip/blob/main/src/open_clip/loss.py).
+
+ModularStarEncoder fine-tuned works with instruction prompts; to get the most out of the model, embed the task in the input. The How to Use section below provides more details.
+
 
 - **Paper:** [Link](arxiv.paper)
 - **Languages:** English, Go, Ruby, Python, Java, C++, PHP, C, JavaScript