FoundationalModels

SageMaker JumpStart Foundation Models - Fine-tuning text generation GPT-J 6B
model on domain specific dataset
Welcome to Amazon SageMaker Built-in Algorithms! You can use SageMaker Built-in algorithms to solve many Machine Learning tasks through SageMaker Python
SDK. You can also use these algorithms through one-click in SageMaker Studio via JumpStart.
In this demo notebook, we demonstrate how to use the SageMaker Python SDK for finetuning Foundation Models and deploying the trained model for inference.
The Foundation models perform Text Generation task. It takes a teyt string as inout and predicts next words in the sequence
How to run inference on GPT-J 6B model without finetuning
How to fine-tune GPT- 6B model on a domain specific dataset, and then run inference on the fine-tuned model. In particular, the example dataset wE demonstrated is publicly available SEC filing of Amazon from ear 2021 to 2022. The expectation is that after fine-tuning, the model should be able to
generate insightful tet in financial domain
We compare the inference result for GPT-J 6B before finetuning and after finetuning Note: This notebook was tested on mI t3 medium instance in Amazon SageMaker Studio with Python 3 (Data Science) kernel and in Amazon SageMaker Notebook
instance with conda nuthon3 kernel