Everyone is talking about: GPTs are custom versions of ChatGPT created by OpenAI users. All you have to do is tell the GPT builder, in plain English, what you want to create, and the builder will take it from there . But how it is actually happened/managed in the backend by OpenAI.
2 Likes
If you need more flexibility, you can use Assistants API.
Thanks for your suggestion @TonyAIChamp 
But I really want to know how this structure actually work.
1 Like
My feeling it is just an average Joe interface to assistants API (hence a lot of settings are pre-defined).
1 Like
Iām a regular user, so my knowledge is incomplete. But from my perspective, this is nothing new, in terms of functionality. Rather, a combination of existing āCustom Instructionsā and āpluginsā functionalities. This is the ability to save multiple named configurations. If you have used the āCustom instructionsā option before, the āCustomized version of ChatGPTā is just an extension of this functionality and the ability to save it as a permanent link that you can share with others. From my perspective, these are different sets of āCustom instructionsā. Very convenient to use, but it is not some super new functionality.
2 Likes
@adam.suskiewicz So, we can say it as ā¦ it is nothing more the instruction tuned version of GPT where the model is able to remember your provided instructions as system prompts do!
Yes, but you also see some backend. You can watch as the AI turns your instructions into an internal prompt. A very educational experience.
Additionally, you can add actions. But as @TonyAIChamp wrote, it is a GUI for functionality that was available via API.
2 Likes
It is another step towards AGI (though nothing ground breaking; just giving higher level functions) It is still in beta. Useful for own tasks, not yet much useful for user-facing apps.
2 Likes
The instructions you entered in the ācreateā tab are transformed into the internal prompt and then you can see the result of this transformation in the āconfigureā tab, in the āInstructionsā field.
1 Like
Why doesnāt anyone mention the fact that files can be given away as the GPT knowledge base ? So the GPT can do some kind of RAG (Retrieval-augmented generation) and integrate the data into its answers. Itās an interesting feature, but I donāt think itās any more advanced than plugins that can read files.
By the way, never upload a file to a public custom GPT that you donāt want users to have access to; itās actually very easy to get access to files. With a little prompt-engineering Iāve been able to download complete PDFs that are copyrighted.
1 Like
The same can be achieved through the standard ChatGPT 4 GUI. I can not see the difference. Files can be uploaded and analyzed, and incorporated into the response text. Itās even more secure than custom GPT, because you didnāt share Your files. The main difference is the repeatability of the queries. In custom GPTs you load once and it remembers it. In the standard GUI you have to reload them in the next session. The difference is preconfiguration, which is very convenient, but, as you mentioned, also dangerous.
import pandas as pd
Load the Excel file to examine its contents
file_path = ā/mnt/data/case_study_SFL.xlsxā
df = pd.read_excel(file_path)Displaying the first few rows of the dataframe to understand its structure
df.head()
1 Like
Much better off not entering instructions into the create tab. Much better and effective for purpose to enter exactly what you want into the configure tab
1 Like
I agree. After using it for some time. At the beginning, it is worth seeing how he transforms himself.
1 Like
If a GPT does not respond as expected, you simply provide the answer you anticipated and ask it what you should enter into its configuration to get the expected answer next time. This works wonderfully. In this way, you use your Custom-GPT to improve itself. However, any changes to the configuration only take effect in subsequent, new chats, a behavior I suggest improving.
3 Likes