Excited to announce a HUGE secret with @LisaSu: @LaminiAI has been building LLMs on @AMD GPUs *in production* for over the past year! We’ve made running LLMs on AMD super easy and a highly competitive option through our LLM Superstation, available now at ~10x lower cost than cloud. 👉🏻 lamini.ai/blog/lamini-amd-pa… Our enterprise customers have already built *thousands* of private LLMs on @LaminiAI LLM Superstations, e.g. @iFit leading at-home fitness with millions of users and @AMD itself: 🚀 Easy & fast: “It was simple to iterate and deploy with a few lines of code and amazingly fast with the AMD Instinct™ hardware.” ⭐️ LLMs are the new IP: “Using a public LLM wasn’t enough: we needed something that we could easily and quickly personalize to our customers’ data and constantly improve on new data, while keeping all of our data private.” ⚙️ Any infrastructure: “We’ve deployed Lamini in our internal Kubernetes cluster with AMD Instinct GPUs, and are using finetuning to create models trained on [our data].” We had a cameo quote from Joe Spisak at @MetaAI who leads the Llama efforts said: “…Llama 2 is becoming the foundation of some of the most innovative companies.” 🫱🏼‍🫲🏾 Join Fortune 500 enterprises, and get your own private LLM Superstations—hosted, VPC, or on-premise (just 2 questions): 9gc3kt44b8q.typeform.com/to/… More (technical) details here👉🏻 lamini.ai/blog/lamini-amd-pa…

I’m super excited to announce @LaminiAI, the LLM engine that gives every developer the superpowers that took the world from GPT-3 to ChatGPT! We make it easy to rapidly train custom LLMs from @OpenAI @EleutherAI @Cerebras @Databricks @HuggingFace @Meta lamini.ai/blog/introducing-l… 🧵

New short course on Fine-tuning LLMs! Many developers are moving beyond only prompting, to also fine-tuning LLMs - that is, taking a pre-trained model and training it further on your own data, which can deliver superior results inexpensively. In this course, @realSharonZhou, CEO of Lamini (disclosure: I’m a minor shareholder) shows you how to recognize when fine-tuning can be help, and how to train an open-source LLM on your own data. I hope you enjoy the course! deeplearning.ai/short-course…

Learn how to improve the accuracy of your LLM apps in our new course with @LaminiAI & @Meta. Taught by experts @realSharonZhou & @asangani7, you’ll learn a development pattern to systematically improve the reliability and accuracy of LLM apps. Join now: hubs.la/Q02Lhdcz0

Excited to share how we’re scaling to thousands of GPUs in production! …with multi-node LLM training, on not just Nvidia but @AMD GPUs Details 👉 lamini.ai/blog/multi-node-ll… Great blog by our team, led by Ayushi 💅 tl;dr - Push the limits of training LLMs on enterprise data scales. - Multi-node training enables data parallelism which speeds up LLM training across multiple nodes and GPUs.

Let's democratize LLMs. Thank you @realSharonZhou and @AndrewYNg for creating a simple and accessible 1-hour course.

Llama 2 on prem🦙 Llama 2 on prem🦙 Llama 2 on prem🦙 Llama 2 on prem🦙 Llama 2 on prem🦙 Llama 2 on prem🦙 lamini.ai/contact

Advancing AI: @LaminiAI Co-founder and CEO @realSharonZhou explains why LLMs are the new IP.

Finetuning your own LLM can solve problems by stopping hallucinations and preventing leakage. Our short course, co-created with @LaminiAI, helps you learn to fine-tune LLMs in a matter of minutes. Learn more about it: hubs.la/Q0226r4W0

Unlike software engineering, prompt engineering requires a unique workflow. In tomorrow’s live workshop, @LaminiAI’s CEO Sharon Zhou will help us demystify prompt engineering for open large language models. Learn more and register here: hubs.la/Q02hd3C70

🧐 Non-fine-tuned LLM vs. Fine-tuned LLM An untrained LLM has no understanding of the world. It is completely random. The first thing we need to do is pre-training. Then, we get a base LLM (non-fine-tuned). After that we can fine-tune the base LLM. The figure shows the procedure. 🖥️ Pre-Training In this phase, we train an LLM with a large amount of data from the internet. The model learns new knowledge in the pre-training step. Pre-training is very expensive and time-consuming. We recommend using a pre-trained LLM and fine-tuning this for your use case. 🚀 Fine-Tuning In the fine-tuning, we optimize an LLM for a specific use case. For example, OpenAI has turned GPT-3 / GPT-4 to ChatGPT. They teach the model to behave more like a chatbot. For this, they used dialog datasets like FAQs, customer support conversations, or Slack messages. We can distinguish two tasks of fine-tuning: - Extraction: We put text in and get less text out (for example summarizing) - Expansion: We put text in and get more text out (for example chatting) In the figure, you can see the difference between a base Llama 2 model and a fine-tuned Llama 2 model (fine-tuned for chatting). We used the Python library @LaminiAI. (@realSharonZhou) You can see that the fine-tuned model gives a comprehensive and correct answer. The non-fine-tuned model answers the question correctly but always repeats itself. 🔎 Useful resource Finetuning Large Language Models  - DeepLearningAI #AI #Lamini #LLM If you enjoyed this post: 👉🏽 Follow us @tinztwins for more Tech and AI content. 👉🏽 Like, Comment and Share the post to support our content 🤝🏽. Thanks. Have a great day! 😉😉

The quote about @AMD's ROCm platform having "software parity" with @nvidia's CUDA platform for large language models came, interestingly, from a former Nvidia CUDA software architect who co-founded the startup.

AI startup Lamini bets future on AMD's Instinct GPUs dlvr.it/SwdKKK

The secret is out. We are ecstatic to see the curtain lifted on @LaminiAI Superstation, powered by AMD Instinct. It's so easy that we are also a customer! The team can't wait to see what Enterprise LLMs developers will tune and personalize with their data. 🤝🤩🌟

🥳Have fun training a tiny free model of your own! Integrated with @streamlit and @LaminiAI . Find the source code in the thread 👇 levelup.gitconnected.com/how…

We had the highest turnout in deeplearning livestream event history! 🎉 Inside joke emoji: 👖 Here's the full recording: piped.video/watch?v=f32dc5M2…

Learn a development pattern to systematically improve the accuracy and reliability of LLM applications in our new short course, Improving Accuracy of LLM Applications, built in partnership with @LaminiAI and @Meta, and taught by Lamini’s CEO @realSharonZhou, and Meta’s Senior Director of Partner Engineering, @asangani7. (Disclosure: I am an investor in Lamini.) The path to tuning an LLM application can be complex. In this course, you'll learn a systematic sequence of steps for improving accuracy by reducing hallucinations: - Create an evaluation dataset to measure model accuracy - Add prompt engineering and self-reflection - Fine-tune your model including "memory-tuning" which is a new method of embedding facts in an LLM Using the Llama 3-8B parameter model, you will: - Build a text-to-SQL agent with a custom schema and simulate situations where it hallucinates - Understand the difference between instruction fine-tuning, which gives pre-trained LLMs instructions to follow, and memory fine-tuning - See how Performance-Efficient Fine-tuning (PEFT) techniques like Low-Rank Adaptation (LoRA) reduce training time by 100x and Mixture of Memory Experts (MoME) reduces it even further I appreciate Meta releasing the Llama's family of open models -- this course gives an example of the unique type of work that developers can do with such models. Please sign up here: deeplearning.ai/short-course…

📣Thrilled to release “Finetune LLMs,” co-created by our CEO @realSharonZhou & @AndrewNg! 👉 Enroll for free now! bit.ly/3siarpD 🥳 Share what you build with us @LaminiAI. We'll showcase the best Lamini llamas (LLMs) with the world!

I've also learned a lot from @AndrewYNg and @DeepLearningAI mini course and from a recent lecture on open source prompt engineering by @realSharonZhou from @LaminiAI: piped.video/live/f32dc5M2Mn0…

You can do the same things as GPT-4 Turbo on every open-source LLM today, @LaminiAI does it all: 🚀 - Structure: Return valid JSON - Speed: Make multiple function calls at once - More knowledge: Retrieval built-in, with finetuning - Longer context: Extend context windows (~128k, using RoPE embeddings) - Ownership: Own the model weights - after all, LLMs are the new IP

Lamini ditches Nvidia in favour of AMD @LaminiAI, an AI startup, is using AMD GPUs instead of the more popular Nvidia GPUs to run large language models (LLMs) like Llama-2 for customers bensbites.beehiiv.com/p/lami…

🤯 Finetuned Question Answering 🤯 Made a small POC on @Replit this morning. Finetuning a LLM with Teslas Q2 2023 earnings report. It's super fast, nimble and accurate in its responses. Demo: replit.com/@homanp/Superagen… A prod ready version will be shipped in Superagent v0.0.1 Under the hood I've used @langchain to retrieve, chunk/split the PDF and @LaminiAI to finetune the model. I'm sold.

Struggling with creating large datasets? 🤯 Lamini augmenters automatically generate high-quality data from <100 examples! 🥳 Install our Python library, augment your dataset, and make training magic today!!🪄 Get started: app.lamini.ai Docs: lamini-ai.github.io/augmente…

Excited to announce a HUGE secret with @LisaSu: @LaminiAI has been building LLMs on @AMD GPUs *in production* for over the past year! We’ve made running LLMs on AMD super easy and a highly competitive option through our LLM Superstation, available now at ~10x lower cost than cloud. 👉🏻 lamini.ai/blog/lamini-amd-pa… Our enterprise customers have already built *thousands* of private LLMs on @LaminiAI LLM Superstations, e.g. @iFit leading at-home fitness with millions of users and @AMD itself: 🚀 Easy & fast: “It was simple to iterate and deploy with a few lines of code and amazingly fast with the AMD Instinct™ hardware.” ⭐️ LLMs are the new IP: “Using a public LLM wasn’t enough: we needed something that we could easily and quickly personalize to our customers’ data and constantly improve on new data, while keeping all of our data private.” ⚙️ Any infrastructure: “We’ve deployed Lamini in our internal Kubernetes cluster with AMD Instinct GPUs, and are using finetuning to create models trained on [our data].” We had a cameo quote from Joe Spisak at @MetaAI who leads the Llama efforts said: “…Llama 2 is becoming the foundation of some of the most innovative companies.” 🫱🏼‍🫲🏾 Join Fortune 500 enterprises, and get your own private LLM Superstations—hosted, VPC, or on-premise (just 2 questions): 9gc3kt44b8q.typeform.com/to/… More (technical) details here👉🏻 lamini.ai/blog/lamini-amd-pa…

Guarantee Valid JSON Output with @LaminiAI smooth ---- Why structured JSON output is so hard 🤔 LLMs are largely based on the transformer architecture, which uses an auto-regressive generator. Transformer treats each word as a token and generates one token at a time. The LLM cannot go back and correct the output once generated, which makes consistent JSON outputs very difficult. ✨ LaminiAI's approach Separating schema generation from model generation using a state machine with support for batching, streaming, and KV cache. As a result, the model still has to process each request individually.