How can we increase Knowledge base of open training model?? can fine tuning do it, because I tried to use openai finetuning It wasn't able to increase knowledgeable using fine-tuning api, so how to feed huge amount of data and increase bot knowledge base

Panchbhai, Harshit 0

I need to increase Knowledge base of open training model. can fine tuning do it, because I tried to use openai finetuning api It wasn't able to increase knowledgeable using fine-tuning api, so how to feed huge amount of data and increase bot knowledge base

romungi-MSFT 42,311 Reputation points Microsoft Employee

2023-03-27T08:16:22.13+00:00
Panchbhai, Harshit Fine tuning will help to improve the model's few-shot learning approach by training the model's weights on your specific prompts and structure. This customization step will let you get more out of the service by providing:

Higher quality results than what you can get just from prompt design

The ability to train on more examples than can fit into a prompt

Lower-latency requests

Are you able to add some valid examples for prompts and completions with at least 200 examples before submitting a fine tuning request?