GenAI

New LLM Pre-training and Post-training Paradigms: A Look at How Moderns LLMs Are Trained

Use cases, techniques, alignment, finetuning, and critiques against LLM-evaluators.

We’re on a journey to advance and democratize artificial intelligence through open source and open science.

Learn how to do batch inference using a provisioned throughput endpoint.

Learn the overview and basic steps for performing model serving on Databricks.

dbdemos - Databricks Lakehouse demos : Fine Tune your LLMs with Mosaic AI Model Training

dbdemos - Databricks Lakehouse demos : LLM Chatbot With Retrieval Augmented Generation (RAG) and DBRX

Discover the power of Lakehouse. Install demos in your workspace to quickly access best practices for data ingestion, governance, security, data science and data warehousing.

Web site created using create-react-app

The easiest & fastest way to run customized and fine-tuned LLMs locally or on the edge

My experiences from preparing for and successfully passing the (beta) exam

Discover why GraphRAG will subsume vector-only RAG and emerge as the default RAG architecture for most use cases.

Educational resources on LLMs

Leverage Azure Cosmos DB for generative AI workloads for automatic scalability, low latency, and global distribution to handle massive data volumes and real-...

GraphRAG, a graph-based approach to retrieval-augmented generation (RAG) that significantly improves question-answering over private or previously unseen datasets, is now available on GitHub. Learn more:

I summarise the kinds of evaluations that are needed for a structured data generation task.

Deep dive into self-improving evaluators in LangSmith, motivated by the rise of LLM-as-a-Judge evaluators plus research on few-shot learning and aligning human preferences.

Read our roadmap on the next wave of enterprise data software in the age of AI.

A practical guide to building successful LLM products, covering the tactical, operational, and strategic.

1. About this blogThis time, I’ll be developing an application designed for use within our FlyersSoft company, to improve workforce efficiency. Idea is to introduce CosmicTalent, an application designed to empower HR and managers in effectively navigating employee information. By leveraging CosmicTalent, users can efficiently filter and identify eligible employees based on specific task requirements. 🚀 Few key takeaways Advanatages of Azure CosmosDB Mongo vCore’s native vector search capabilities over Azure Vector Search.

Machine learning, Scala, and interactive computing

Samples to demonstrate pathways for Retrieval Augmented Generation (RAG) for Azure Data - microsoft/AzureDataRetrievalAugmentedGenerationSamples

REFERENCES: 1. Build an LLM from Scratch book: https://mng.bz/M96o 2. Build an LLM from Scratch repo: https://github.com/rasbt/LLMs-from-scratch 3. Slides: https://sebastianraschka.com/pdf/slides/2024-build-llms.pdf 4. LitGPT: https://github.com/Lightning-AI/litgpt 5. TinyLlama pretraining: https://lightning.ai/lightning-ai/studios/pretrain-llms-tinyllama-1-1b DESCRIPTION: This video provides an overview of the three stages of developing an LLM: Building, Training, and Finetuning. The focus is on explaining how LLMs work by describing how each step works. OUTLINE: 00:00 – Using LLMs 02:50 – The stages of developing an LLM 05:26 – The dataset 10:15 – Generating multi-word outputs 12:30 – Tokenization 15:35 – Pretraining datasets 21:53 – LLM architecture 27:20 – Pretraining 35:21 – Classification finetuning 39:48 – Instruction finetuning 43:06 – Preference finetuning 46:04 – Evaluating LLMs 53:59 – Pretraining & finetuning rules of thumb

Here’s everything you need to know to build your first LLM app and problem spaces you can start exploring today.

Pliny the Prompter has been finding ways to jailbreak, or remove the prohibitions and restrictions on leading LLMs, since last year.

Explore the new LangChain RAG Template with Redis integration. Streamline AI development with efficient, adaptive APIs.

Build your own AI copilot with vCore-based Azure Cosmos DB for MongoDB and Azure OpenAI.

Creating your own AI agent equipped with a sophisticated memory system. This guide provides a detailed walkthrough on leveraging the capabilities of Fireworks AI, MongoDB, and LangChain to construct an AI agent that not only responds intelligently but also remembers past interactions.