Content
@
0 reply
0 recast
0 reaction
tldr (tim reilly)
@tldr
a "Farcaster Lore" packet for Agent training context window.
5 replies
3 recasts
39 reactions
KMac🍌 ⏩ ツ
@kmacb.eth
I wonder if @adrienne’s RAG pipeline with GM FC content gets you there
1 reply
0 recast
4 reactions
tldr (tim reilly)
@tldr
GMFC would be absolutely perfect. What's RAG pipeline?
3 replies
0 recast
2 reactions
shoni.eth
@alexpaden
semantic search! pinecone database of a summary where you search the embedding for semantically similar content RAG is just the process of search + inject the results into your prompt
1 reply
0 recast
2 reactions
shoni.eth
@alexpaden
not just summary but any content. i.e a book is broken into chunks and each chunk is embedded (an array of 1536 float 32s for example) that embedding array is your search key/id. the metadata might be the page number
1 reply
0 recast
2 reactions
