Jan 11, 2024 / Chatbots (experimental), Filter by Rich Text snippets in Views, shorter trial & no free version

[mdubakov]

@pzol

can we have variable height rows? Can the row height setting serve as a maximum height instead of always taking up the full height??

This is challenging to implement, table can jump after fully load, so it is quite hard to make it look pretty.

[cannibalflea]

Can we have the height setting per database in the table view when we have a hierarchy? Having text snippets is very useful but sometimes the higher rows don’t need the same space, so we end up wasting unneeded space.

[mdubakov]

@Oshyan

After 24hrs both of my added Chatbots still show “Reading”/yellow light status. I may have tried to create too many of them close together (I made 3 over the course of an hour or so), not realizing that indexing took longer than I thought. But it does seem like things are stuck here since since my OpenAI key usage has not changed since yesterday. And unlike the AI Search Indexing there is no way to reindex a chatbot as far as I can see, you have to just remove and recreate it, which is unfortunate

We’re looking into it. It seems all index are good, but we will improve visual indicator in future.

[interr0bangr]

Which version of ChatGPT is being used for the ChatBot feature?

When using automations, I know we can specify what version we want to use like this:

[ai temperature=0.5 maxTokens=4000 model=gpt-4-1106-preview])

but I don’t see any options like that on the AI settings page.

[mdubakov]

Chatbots use GPT4. So far there is no option to setup it.

[interr0bangr]

Great, thanks. Not looking to change it, was just interested in knowing what you using!

[Yuri_BC]

The chatbot does not have direct access to, or cannot interpret, the metadata associated with entities in Fibery. It fails to fetch specific details like the name, database, and creation date of an entity. This is confirmed by the Fibery chatbot its following answers:

"Based on the provided context and chat history, while I do possess knowledge and am aware of titles or names, I may not be able to associate or distinguish if a certain piece of knowledge belongs to a resource with a specific title or name. This is because the AI’s understanding and retrieval of information depends on the specific structure and organization of the data in Fibery, as well as the AI’s capabilities. "

and

“As an AI, I can recognize and understand the titles or names of resources, and I do not make assumptions about this information. However, associating these names with specific containers of knowledge, such as the fields of the entities with these names, requires additional context or metadata. This is because my ability to make such associations depends on the specific structure and organization of the data in Fibery, as well as my AI capabilities.”

So what I understand is that the connection between Chatbot knowledge and knowledge containers (like rich text fields, other rich text fields of the same entity and their parent entity) could only be available if each rich text field individually contains metadata about itself and its entity and fields?