sharonzhou

ROCm is great. We were able to get run and finetune LLMs on AMD Instincts with parity to NVIDIA A100s - and built an SDK that’s as easy to use as HuggingFace or easier (Lamini). Or at the very least, our designer is able to finetune/train the latest LLMs on them like Llama 2 - 70B and Mistral 7B with ease. The ROCm library isn’t as easy to use as CUDA because as another poster said, the ecosystem was built around CUDA. For example, it’s even called “.cuda()” in PyTorch to put a model on a GPU, when in reality you’d use it for an AMD GPU too.

Can do training here (for free on small LLMs). With a simple interface.

https://colab.research.google.com/drive/1QMeGzR9FnhNJJFmcHtm...

https://lamini-ai.github.io/

from llama import QuestionAnswerModel

model = QuestionAnswerModel()

model.load_question_answer_from_csv("data.csv")

model.train() # returns id to run inference & playground interface

Official repo for the finetuning pipeline, so you can train custom models on your data.

It's free, on small LLMs It's fast, taking 10-15 minutes It's like working with an unlimited prompt size, with 1000x+ more space than the largest prompts It's learning new information, not just trying to make sense of it given what it already learned (retrieval-augmented generation)

Layperson's blogpost: https://www.lamini.ai/blog/free-fast-and-furious-finetuning

Hi HN!

I’m super excited to announce Lamini, the LLM engine that gives every developer the superpowers that took the world from GPT-3 to ChatGPT!

I’ve seen a lot of developers get stuck after prompt-tuning for a couple days or after fine-tuning an LLM and it just gets worse—there’s no good way to debug it. I have a PhD in AI from Stanford, and don’t think anyone should need one to build an LLM as good as ChatGPT. A world full of LLMs as different & diverse as people would be even more creative, productive, and inspiring.

That’s why I’m building Lamini, the LLM engine for developers to rapidly customize models from amazing foundation models from a ton of institutions: OpenAI, EleutherAI, Cerebras, Databricks, HuggingFace, Meta, and more.

Here’s our blog announcing us and a few special open-source features! https://lamini.ai/blog/introducing-lamini

Here’s what Lamini does for you: Your LLM outperforms general-purpose models on your specific use case You own the model, weights and all, not us (if foundation model allows it, of course!) Your data helps the LLM, and build you an AI moat Any developer can do it today in just a few lines of code Commercial-use-friendly with a CC-BY license

We’re also releasing several tools on Github: Today, you can try out our hosted data generator for training your own LLMs, weights and all, without spinning up any GPUs, in just a few lines of code from the Lamini library. https://github.com/lamini-ai/lamini/

You can play with an open-source LLM, trained on generated data using Lamini. https://huggingface.co/spaces/lamini/instruct-playground

Sign up for early access to the training module that took the generated data and trained it into this LLM, including enterprise features like virtual private cloud (VPC) deployments. https://lamini.ai/contact

I was a couple months ago, then I started my own thing and I feel great again. It’s called Sidepact and it’s actually really relevant to this problem! It’s a weekend program for employed engineers to start companies—to get people out of their local maxima. Check it out and apply if what you’re looking for next is to maximize fulfillment in your work and you want to own it :) http://sidepact.com

I think it's less about complexity and more about precision. A simple law can be extremely difficult to encode if it lacks specification.

I'm surprised there was no octopus to represent the bandits.

In all seriousness, this post makes sense to me, as someone who does RL research. However, the intuition behind the concepts could be communicated more clearly. I would reason that this piece is less accessible to those who have much less knowledge of RL/bandits. Given that it's an introduction, I presume that's your intended reader, though perhaps writing can also be for your own edification. Who's your audience?

Sounds like she knows her priorities and adamantly avoids wasting money and chasing things for reputation's sake. I aspire to be like that.

This looks really cool and relevant. Following.