← Graph

Large Language Models

concept 7 connections

Deep artificial neural networks with general-purpose language understanding and generation capabilities. Exploded in popularity after the 2017 Google paper 'Attention Is All You Need' which introduced the Transformer architecture. Strong at structuring unstructured data (paragraph → JSON/YAML), summarization, classification, language translation, content generation, and chatbot-style Q&A. Models are trained on a snapshot in time (e.g. GPT-4 through April 2023), so data becomes stale and they can hallucinate.

category
architecture
talk Building LLM powered applications in Ruby
about
Large Language Models concept
Explains LLMs and their strengths.
talk Next Token!
about
Large Language Models concept
Core subject — debunking how LLMs actually work.
takeaway LLMs Are Just Token Generators
about
Large Language Models concept
Reframes LLMs as next-token predictors wrapped in scaffolding.
question LLMs 10× product misfunctions too
about
Large Language Models concept
Question is about LLM impact on product quality.
talk Discipline Doesn't Scale
about
Large Language Models concept
LLMs multiply the number of contributions and stress the automation–contribution gap.
takeaway LLMs Amplify Existing Patterns
about
Large Language Models concept
Takeaway specifically about LLM impact on software quality.
concept Large Language Models
related_to
Transformers concept
LLMs are built on the Transformer architecture.

Provenance

Created in
Building LLM-Powered Applications in Ruby — Andrei Bondar... 2026-04-17 23:20
Read by
7 extractions