Activity
Mon
Wed
Fri
Sun
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
What is this?
Less
More

Memberships

AI Automation Society

208.5k members • Free

1 contribution to AI Automation Society
AI does not find data in the Rag database.
Hello everyone, I'm using Nate's template from this video: https://www.youtube.com/watch?v=Kvf-Pr1LRe4&t=905s I managed to make it work with a small file, but when placing a 13-page file, it was unable to find the answer, as I imagine that only a portion is being sent to the AI ​​to check if it finds what I need, how can I increase the loop, in that the AI ​​will receive more data to look for what I need? In the logs I can see that only a portion was sent to AI, and no new attempts were made with the rest of the document's content. I have basically the same settings as Nate, just logged in with my credentials and a few files on the drive.
0 likes • Jan 21
I managed to get it to work by increasing the Vector Store Tool limit to 20. But what will happen when I have 150 documents or more? Will I need to put this infinite number? Where will IA look until he finds it, spending thousands of tokens and taking a long time to respond? I tried changing the embeding to large, but when running the workflow I encounter errors, such as: Error inserting: expected 1536 dimensions, not 3072 400 Bad Request on Insert into Supabase Vectorstore. So it's working, but just on small files, If anyone has a cool solution, I'll gladly accept it.
1-1 of 1
Gabriel Delazeri
1
4points to level up
@gabriel-delazeri-2274
Hi aasdasdasdasdasddddddddddddddddddddd

Active 314d ago
Joined Jan 20, 2025
Powered by