Getting killer results from your PickAxe

General
1

Over the past week, to make the knowledge base and the role instructions function better - I have used the projects inside of GPT to do two things.

  1. I use ChatGPT to extract a concepts CSV from my documents and build a knowledge base. This does a similar things as the upload - but it gives me better control. I tell it to create a category / sub-category and concepts.
  2. Review the CSV for missing items. Add items that are missing. For example - it missed a major concept - so I simply ask ChatGPT to build more concepts based on what it missed.
  3. Upload the CSV to PickAxe - I typically separate out CSVs by major “Core Concepts” as shown below in the instructions.
  4. Then back in ChatGPT, based on the concepts CSV - I ask it the following “let’s write detailed instructions for the Assistants API. Format it in plain text with no markup to be consumed by the API. Include the following: Role, Knowledge Base & Core Concepts, Rules, Interaction Flow, Output.”

I am getting killer results, and adding new knowledge base - I can just edit a few areas (Knowledge Base & Core Concepts, and Interaction Flow).

Let me know if you try this and how you improve on what I have done here.

11 Likes
2

Could you give an example of what one concept might be to help us understand the csv better. I’m not fully understanding TY.

3

A concept is what happens when you upload the doc - then look at the breakdown of that DOC (in RAG) in the PickAxe learn. My method - you get more control over the concepts that are being extracted from the doc - and make sure it is complete. Sometimes it doesn’t catch things, or at least not the way you would suggest.

2 Likes
4

Thank you for the extra clarity.

5

@b2bea that’s a clever way of creating the knowledge base.

(For those not aware of this feature in Pickaxe, CSV files are treated differently than other files - see this article to better understand the difference: Putting a large spreadsheet (csv) in the Knowledge Base - #4 by admin_mike)

6

Great stuff! Still getting my head around it but I will try it out!

7

I followed the instructions above about the concepts and had the Pickaxe builder build the bot with that information.
The bot still only retrieves 3-5 chunks.
My CSV has 64,892. What is the limit? I had separate files earlier, and had the same problem.
Do you have a live chat that can help me?

8

When you go into the Knowledge Base and see your CSV in there, how many chunks does it say are in there? Does it say 4 or does it say 64,892?

If it only says 4, then your entire spreadsheet was not ingested. If it says ~64,000, then you should go down and increase the Amount that it pulls from the KB.

Screenshot 2025-02-25 at 2.46.02 PM
Screenshot 2025-02-25 at 2.46.02 PM1568×956 65.7 KB

1 Like
9

Hi Mike, the chuck shows ~64,000. I adjusted the amount section, but that only has numbers up to 25,000. Maybe that will do? I will test now.

Screen Shot 2025-02-25 at 2.58.37 PM
Screen Shot 2025-02-25 at 2.58.37 PM1476×352 20.7 KB

Screen Shot 2025-02-25 at 3.00.49 PM
Screen Shot 2025-02-25 at 3.00.49 PM1414×498 18.4 KB

10

I just tested it after setting up the amount to 25,000.
The spreadsheet is for home sales data in Santa Cruz County, California. Each line is for separate address sold. Maybe it’s something with the formatting? Do you have any other suggestions?

11

It seems to not even recognize the data with dates from last year. It says it doesn’t have it, even though it is in the spreadsheet.

12

Hi @maaret I recently had the same issue while working on a clients project. The fox was quite simple. Delete the CSV, then reupload.

Most importantly, utilize the ‘Knowledge explorer’ feature to input test queries and edit the knowledge chunks according to your project goals.

Lastly, to ensure killer results from your Pickaxe, you want to ensure that your overall settings are properly tweaked. Here’s an in depth tutorial I made for your reference: