Link: https://textfx.withgoogle.com/
How can an LLM be trained to perform things like this experiment site does? (Visit the site and you will find out more about these tools.) This is different from my previous question about LLM tasks because it talks about this experiment's features.
TextFX is open source and every example has a "look behind the scenes" button that shows the prompt, which is broken down into a preamble and then a set of examples:
For instance, their prompt for the acronym generator is:
An acronym is an abbreviation of several words in such a way that the abbreviation itself forms a pronounceable word.
Here is an acronym that uses the letters of the word "rap": RAP - Recognizing Analogous Patterns
Here is an acronym that uses the letters of the word "mural": MURAL - Magnifying Urban Realities Affecting Lives
[continued] Here is an acronym that uses the letters of the word "":
They then do postprocessing to check that the completion matches the expected format.
