Add LLM FlowerTune

doc/source/ref-changelog.md

@@ -16,6 +16,10 @@
 
   Flower Modifiers (we call them Mods) can intercept messages and analyze, edit or handle them directly. Mods can be used to develop pluggable modules that work across different projects. Flower 1.8 already includes mods to log the size of a message, the number of parameters sent over the network, differential privacy with fixed clipping and adaptive clipping, local differential privacy and secure aggregation protocols SecAgg and SecAgg+. The Flower Mods API is released as a preview, but researchers can already use it to experiment with arbirtrary SMPC protocols.
 
+- **Fine-tune LLMs with LLM FlowerTune** ([#3029](https://github.com/adap/flower/pull/3029), [#3089](https://github.com/adap/flower/pull/3089), [#3092](https://github.com/adap/flower/pull/3092), [#3100](https://github.com/adap/flower/pull/3100), [#3114](https://github.com/adap/flower/pull/3114), [#3162](https://github.com/adap/flower/pull/3162), [#3172](https://github.com/adap/flower/pull/3172))
+
+  We are introducing LLM FlowerTune, an introductory example that demonstrates federated LLM fine-tuning of pre-trained Llama2 models on the Alpaca-GPT4 dataset. The example is built to be easily adapted to use different models and/or datasets. Read our blog post [LLM FlowerTune: Federated LLM Fine-tuning with Flower](https://flower.ai/blog/2024-03-14-llm-flowertune-federated-llm-finetuning-with-flower/) for more details.
+
 - **Introduce built-in Differential Privacy (preview)** ([#2798](https://github.com/adap/flower/pull/2798), [#2959](https://github.com/adap/flower/pull/2959), [#3038](https://github.com/adap/flower/pull/3038), [#3147](https://github.com/adap/flower/pull/3147), [#2909](https://github.com/adap/flower/pull/2909), [#2893](https://github.com/adap/flower/pull/2893), [#2892](https://github.com/adap/flower/pull/2892), [#3039](https://github.com/adap/flower/pull/3039), [#3074](https://github.com/adap/flower/pull/3074))
 
   Built-in Differential Privacy is here! Flower supports both central and local differential privacy (DP). Central DP can be configured with either fixed or adaptive clipping. The clipping can happen either on the server-side or the client-side. Local DP does both clipping and noising on the client-side. A new documentation page [explains Differential Privacy approaches](https://flower.ai/docs/framework/explanation-differential-privacy.html) and a new how-to guide describes [how to use the new Differential Privacy components](https://flower.ai/docs/framework/how-to-use-differential-privacy.html) in Flower.
@@ -44,7 +48,7 @@
 
   As usual, we merged many smaller and larger improvements to the documentation. A special thank you goes to [Sebastian van der Voort](https://github.com/svdvoort) for landing a big documentation PR!
 
-- **General updates to Flower Examples** ([#3029](https://github.com/adap/flower/pull/3029), [3134](https://github.com/adap/flower/pull/3134), [3114](https://github.com/adap/flower/pull/3114), [3089](https://github.com/adap/flower/pull/3089), [2996](https://github.com/adap/flower/pull/2996), [3092](https://github.com/adap/flower/pull/3092), [3100](https://github.com/adap/flower/pull/3100), [2930](https://github.com/adap/flower/pull/2930), [2967](https://github.com/adap/flower/pull/2967), [2467](https://github.com/adap/flower/pull/2467), [3172](https://github.com/adap/flower/pull/3172), [2910](https://github.com/adap/flower/pull/2910), [3162](https://github.com/adap/flower/pull/3162), [#2918](https://github.com/adap/flower/pull/2918), [#2773](https://github.com/adap/flower/pull/2773), [#3063](https://github.com/adap/flower/pull/3063), [#3116](https://github.com/adap/flower/pull/3116), [#3117](https://github.com/adap/flower/pull/3117))
+- **General updates to Flower Examples** ([3134](https://github.com/adap/flower/pull/3134), [2996](https://github.com/adap/flower/pull/2996), [2930](https://github.com/adap/flower/pull/2930), [2967](https://github.com/adap/flower/pull/2967), [2467](https://github.com/adap/flower/pull/2467), [2910](https://github.com/adap/flower/pull/2910), [#2918](https://github.com/adap/flower/pull/2918), [#2773](https://github.com/adap/flower/pull/2773), [#3063](https://github.com/adap/flower/pull/3063), [#3116](https://github.com/adap/flower/pull/3116), [#3117](https://github.com/adap/flower/pull/3117))
 
   Two new examples show federated training of a Vision Transformer (ViT) and federated learning in a medical context using the popular MONAI library. `quickstart-pytorch` and `quickstart-tensorflow` demonstrate the new Flower Next `ServerApp` and `ClientApp`. Many other examples received considerable updates as well.