AWS Workshop (SF): Fine-Tune and Host SLMs (Llama3 or Mistral)


Sep 04, 10:00 AM PDT
  • US-San Francisco (AWS Loft SF, 525 Market Street San Francisco, C...) AWS
  • 173 RSVP
Description
Speaker

How can technology leaders navigate the complexity of fine-tuning AI models for improved performance while also balancing costs of training and deployment?

Join this technical deep-dive with AI experts from AWS, Meta to learn how to easily fine-tune and host their own, task-specific small LLMs (SLM) like Llama3 or Mistral. We will walk you through best practices, solution accelerators and pre-built examples (free access post workshop)

Agenda:
* 10:00am~10:30am: Checkin and networking
* 10:30pm~12:00pm: Tech talks and hands-on workshops
* 12:00pm~1:00pm: Lunch break
* 13:00pm~2:30pm: Tech talks and hands-on workshops
* 14:30pm~3:00pm: Open discussion and mingle

Why Attend?

  • Understand SLMs: Learn about small language models (SLMs), which typically have 100 million to 1 billion parameters, offering a compact yet powerful alternative to large language models (LLMs).
  • Maximize Accuracy: Discover how SLMs can outperform LLMs in task-specific generative AI applications.
  • Cost Efficiency: Save over 90% on inference costs, making SLMs ideal for heavy inference use-cases with reduced compute demand and increased accessibility.
  • Architectural Advancements: Explore the latest in data multi-tenancy with Adapters, ensuring robust data management and dynamic query routing.
  • Optimized Fine-Tuning: Gain insights into avoiding data over-fitting or under-fitting during fine-tuning, and understand the differences in benchmarks when fine-tuning.
  • Customer Experience: Learn how to improve generative AI customer questions and feedback experiences using data training techniques.

    • Exclusive Benefits

  • Personalized Guidance You’ll get 1:1 workshop assistance with AWS and Invisible experts.
  • Enhanced Support Access AWS-funded SI Partner engagement for data preparation and fine-tuning assistance.
  • AWS Credits Receive AWS credits for workload migration and an additional $150K credits for AWS silicon users.
  • Post-Workshop Resources Gain free access to solution accelerators and pre-built examples to implement immediately.
  • SLM Adoption Program Learn about the AWS programs for SLM adoption, providing comprehensive support and resources.

    • Speakers:
    - Deepam Mishra and Adam Naor, AWS Startups
    - Clement Perrot, Ganapathi Krishnamoorthi, Dmitry Soldatkin and Marco Punio, AWS
    - Lydia Anderson, Tiarne Hawkins, Invisible Technologies
    - In addition, we will have 10+ technical and business leaders from AWS and Meta.

    Requirements:
    Bring your laptop

    Venue:
    AWS Loft | San Francisco, 525 Market Street, 2nd Floor, San Francisco, CA 94105
    Main entrance is up the stairs to SW of building next to Chipotle restaurant. Alternative access is using elevators in lobby.
    Check Slack/Discord for more driving instructions, parking info, public transportations, car pool, etc..

    *This event is hosted by our partner, AWS and Invisible.

    Sponsors:
    We are actively seeking sponsors to support AI developers community.  Whether it is by offering venue spaces, providing food, or cash sponsorship. Sponsors will not only speak at the meetups, receive prominent recognition, but also gain exposure to our extensive membership base of 30,000+ AI developers in San Francisco Bay Area or 400K+ worldwide.

    Community on Slack/Discord
    - Event chat: chat and connect with speakers and attendees
    - Sharing blogs, events, job openings, projects collaborations
    Join Slack (search and join the #sanfrancisco channel) | Join Discord

    Deepam Mishra (AWS), Clement Perrot (AWS), +

    The event ended.
    Watch Recording
    *Recordings hosted on Youtube, click the link will open the Youtube page.
    Contact Organizer