feat: enable code llama finetuning in JumpStart (#4586)

* code llama finetuning * code llama finetuning --------- Co-authored-by: Xin Huang <xinxh@amazon.com>
aws · Mar 11, 2024 · cc112d5 · cc112d5
1 parent 8da3ca8
commit cc112d5
Show file tree

Hide file tree

Showing 2 changed files with 692 additions and 0 deletions.
diff --git a/introduction_to_amazon_algorithms/jumpstart-foundation-models/README.md b/introduction_to_amazon_algorithms/jumpstart-foundation-models/README.md
@@ -6,6 +6,7 @@ These examples provide quick walkthroughs to get you up and running with Amazon
 
 - [JumpStart Text Generation Inference](jumpstart-text-generation-inference.ipynb) provides a single notebook for all JumpStart text generation models using default payloads retrieved from the JumpStartModel.
 - [Llama-2 Text Generation](llama-2-text-completion.ipynb) and [Llama-2 Chat Completion](llama-2-chat-completion.ipynb) demonstrate text generation using Meta's Llama-2 pretrained and fine-tuned chat models, respectfully.
+- [Code LIama fine-tuning](code-llama-fine-tuning-evaluate-human-eval.ipynb) demonstrate Code Llama fine-tuning and evaluation using [human-eval repository](https://github.com/openai/human-eval).
 - [Llama Guard Text Moderation](llama-guard-text-moderation.ipynb) demonstrates using a Llama Guard model to moderate a conversation in conjunction with a Llama-2 model.
 - [Gemma Text Generation Fine-tuning](gemma-fine-tuning.ipynb) demonstrates fine-tuning and deployment on Google's Gemma text generation models in SageMaker JumpStart.
 - [Text2Text Generation Flan T5 UL2](text2text-generation-flan-t5-ul2.ipynb) demonstrates Text2Text generation using state-of-the-art pretrained Flan T5 models from [Hugging Face](https://huggingface.co/docs/transformers/model_doc/flan-t5) which take an input text containing the task and returns the output of the accomplished task.