aqader

this is really cool, can’t wait to try it out for some ML pipeline development. kudos myles and akshay!

Yes, we did! The dataset has since been cleaned even more so we're due to update the model.

There are a couple open source implementations. I'll list a couple below:

7B: - https://huggingface.co/tloen/alpaca-lora-7b - https://huggingface.co/ozcur/alpaca-native-4bit 13B: - https://huggingface.co/samwit/alpaca13B-lora - https://huggingface.co/Dogge/alpaca-13b 30B: - https://huggingface.co/baseten/alpaca-30b - https://huggingface.co/Pi3141/alpaca-30B-ggml

Depends on the model size. A model like GPT3 that has hundreds of billions of paramaters, you can do few-shot learning with. You'll still pay for the tokens processed and it'll at least linearly increase response times the larger your input is.

Fine-tuning can get you similar results on smaller / faster models. The downside is you have to craft the dataset in the right way. There are trade-offs to both approaches but fwiw, I don't think Alpaca-7b can do few-shot learning.

Almost. If your dataset contains questions and answers about your own projects documentation, then yes. The UX around how to prompt a fine-tuned model depends on the format of the dataset it's trained on.

One way you can do this is pass your documentation to a larger model (like a GPT3.5 / OSS equivalent) and have it generate the questions/answers. You can then use that dataset to fine-tune something like Llama to get conversation / relevant answers.

Yeah, this is running in 8bit mode. The 30b 8bit version we released seems to do a lot better but it requires significantly more compute.

https://huggingface.co/baseten/alpaca-30b

For this demo, we're using the 8bit version here: https://huggingface.co/tloen/alpaca-lora-7b

We also fine-tuned and OSS'd a 30b version here that you can checkout (on the cleaned 52k Alpaca dataset) https://huggingface.co/baseten/alpaca-30b