Our great sponsors
-
dolly
Databricks’ Dolly, a large language model trained on the Databricks Machine Learning Platform
-
WorkOS
The modern identity platform for B2B SaaS. The APIs are flexible and easy-to-use, supporting authentication, user identity, and complex enterprise features like SSO and SCIM provisioning.
I'm going through the dataset with your datasette tool and it looks like it might be a good idea to clean things up a bit. There are many duplicates[1], creepypastas[2] and other strange things in there.
[1] https://lite.datasette.io/?json=https%3A%2F%2Fraw.githubuser...
[2] https://lite.datasette.io/?json=https://github.com/databrick...
can you compare your dolly offering with https://github.com/microsoft/DeepSpeedExamples/blob/master/a...
it's probably simple for Dolly v1 (?) since it was a fine-tuned version of GPT-J
https://github.com/ggerganov/ggml/tree/master/examples/gpt-j
AFAIK there is no .cpp version of Pythia-12B yet
I saw this: https://github.com/jankais3r/LLaMA_MPS
it runs slightly slower on the GPU than under llama.cpp but uses much less power doing so
I would guess the slowness is due to immaturity of the PyTorch MPS backend, the asitop graphs show it doing a bunch of cpu along with the gpu, so it might be inefficiently falling back to cpu for some ops and swapping layers back and forth (I have no idea, just guessing)
Related posts
- Apples effort on developing Chat GPT like functions?
- A full tutorial on turning GPT-2 into a conversational AI
- NanoChatGPT - turning nanogpt into a chat model/LLM
- Introducing Lam: is an all-in-one desktop open-source app built with Electron and React. It allows users to perform various tasks by installing and using different apps within the main app. The chat interface with LLM enables seamless interaction with the installed apps.
- Lam: All-in-one open-source desktop app