Content
@
https://warpcast.com/~/channel/aichannel
0 reply
0 recast
0 reaction
Stephan
@stephancill
Is there an easy way to fine tune an LLM on stuff I write so that I can talk to myself? Transforming data is not a requirement, can do it myself
7 replies
4 recasts
25 reactions
shoni.eth
@alexpaden
it’s mostly fine tuned on your style, and rag for historical data. depending on the granularity of data, you’ll have diminishing returns on fine tuning
1 reply
0 recast
1 reaction
Stephan
@stephancill
Maybe both then. I’m interested in retaining style
1 reply
0 recast
1 reaction
shoni.eth
@alexpaden
if you're doing both i'll suggest starting with rag because it's easier, can guide training data for fine tuning, and because it will pick up on some styling just from the rag results. pinecone, pgvec, and sqlitevec all work well + a lot of the demo repos now are basically one stop shops for dumping text in a folder and spinning rag up from what i've seen (haven't used them)
0 reply
0 recast
0 reaction
