Viewing Text Extractor Output

While you can view documents you've loaded into a Prompt Studio project in the Doc View tab, it is not this document that's passed on to the LLM. Any input document is first send to a Text Extraction service so that raw text is extracted from it. That text is then further processed and sent to the LLM. As a result, the quality of the raw text extraction is very important. Like other things in computing, even for LLMs: garbage in, garbage out.

Below is a side-by-side comparison of Doc View and Raw View.

Debugging with Raw View

If a prompt response is wrong, one of the very first places you should check is the Raw View to ensure that the output to the LLM is indeed correct. Switch to Raw View and check portions of the input document to see why a prompt response might be the way it is. Especially in OCR mode, a lot of mistakes can be made with text extraction even in the cleanest looking documents.