add more details for peft

examples/advanced/llm_hf/README.md

@@ -2,8 +2,12 @@
 This example shows how to use [NVIDIA FLARE](https://nvidia.github.io/NVFlare) for Large Language Models (LLMs) tuning tasks. It illustrates how to adapt a local training script with [HuggingFace](https://huggingface.co/) trainer to NVFlare.
 
 ## Introduction 
-This example illustrates both supervised fine-tuning (SFT) and parameter-efficient fine-tuning (PEFT) using the [SFT Trainer](https://huggingface.co/docs/trl/sft_trainer) from [HuggingFace](https://huggingface.co/).
+This example illustrates both supervised fine-tuning (SFT) and parameter-efficient fine-tuning (PEFT) using the [SFT Trainer](https://huggingface.co/docs/trl/sft_trainer) from [HuggingFace](https://huggingface.co/) with [PEFT library](https://github.com/huggingface/peft).
+
 We used the [Llama-2-7b-hf model](https://huggingface.co/meta-llama/Llama-2-7b-hf) to showcase the functionality of federated SFT and PEFT, allowing HuggingFace models to be trained and adapted with NVFlare. 
+
+For PEFT, we used LoRA method, other PEFT methods (e.g. p-tuning, prompt-tuning) can be easily adapted as well by modifying the configs following [PEFT](https://github.com/huggingface/peft) examples.
+
 Mainly on two fronts:
 - Adapt local HuggingFace training scripts to federated application
 - Handling large model weights (~26 GB for Llama-2-7b-hf model), this is supported by NVFlare infrastructure, and does not need any code change.