fine-tuning gpt on custome knowledge base
Done - api call -- 3.5 turbo and 4 successful #ERROR - response breaks in half leanth size issue (error with multiple call)
#Done - Basic Frontend ##adding like and dislike button ##adding backend log file ##getting where the model fails
get user manul (if there are any) Srijan
In progress - Threading ( crntly all converation in one thread )
Fun to do : langchin introduction an play with more parameter
future: verctor embedding function calling
data :: kaggle
model : hugging face
generation : google collb
find alternative's for all ur googel product in mcirosoft