This repository will be used to share my non-sensitive LLMs work for educational purposes.
A common real world use case is to recognize the presence of a phrase in text. For example, in the manufacturing context, there may be text descriptions to describe the suspected cause for a product failure. Analyzing such data can be useful for the manufacturer to identify common causes of failures and take steps to address these common causes. For more sophisiticated applications, data related to the products can be used to correlate to these causes of failure and machine learning models can be employed to predict these causes.
However, the main challenge is identifying key descriptions in the data. This notebook gives a high level illustration of how we can coax LLMs to identify key descriptions in the data. While this is only an example, these snippet of codes can be extended to for more sophisticated applications.