← 2 AI (models)


Demos

  • token interaction
  • attention
  • sequence routing
  • FFNs


This page describes

  • 2.3.1 Gist of LLMs. The gist of what an LLM model does without complex algorithms or architecture diagrams.
  • 2.3.2 Gist of the LLM TF UFA. A marvel of engineering complexity.
  • 2.3.3 Agentic
  • 2.3.4 LLM transformer training
  • 2.3.5 LLM Internal Agent. My own invented term. iAgent + TF = LLM.
  • 2.3.6 LLM stack demos. Listed on my wiki.

drones


2.3.1 Gist of LLMs

Describes in concise detail gist of what an LLM model does without describing complex algorithms or architectures.

smol


2.3.2 Gist of the LLM TF UFA (GPT-3)

This will give you an appreciation of the vast complexity of TF calculations.

01


2.3.3 Agentic LLMs


2.3.4 LLM transformer training

Training is something most of us will never do (Palantir uses standard models; they do no customized training). But Training is the defining aspect of model transformer (TF) design.

smol


2.3.5 LLM Internal Agent

The term “agent” normally means a deterministic (non-AI, not GPU-based) control loop that is the “caretaker” or interface between the LLM model and the outside world. However, there is also an agent (what I call an internal agent or “iAgent”) in the LLM that is the interface between the transformer (TF) and the outside (of the LLM) world. Again, as with training, understanding this helps to understand the core nature of what an LLM and an agent is. An (external) agent is to some extent just an extension or a partner of the internal agent.

iagent


2.3.6 LLM stack demos

My focus at one point was on LLMs. I thought that that was going to be the market focus. There is a lot of value in being able to deploy and fine-tune your own LLMs, but for most of us these skills will not be used often.

The wiki page “AI LLM stacks” (a bit chaotic after a recent wiki reorganization) lists the following subpages are of interest:

smol


26.0524