AI & machine learning2023 · shared on Hacker News

TinyStories: How Small Can Language Models Be and Still Speak Coherent English? (2023)

Explore TinyStories: How Small Can Language Models Be and Still Speak Coherent English? (2023), its original source and Hacker News discussions. Discover community perspectives and save it to your reading list.

The conversation around the idea.

Links found in the discussion

OpenAI says it has evidence DeepSeek used its model to train competitor

What's the strongest AI model you can train on a laptop in five minutes?

LLM from scratch, part 28 – training a base model from scratch on an RTX 3090

Phi-4: Microsoft's Newest Small Language Model Specializing in Complex Reasoning

“Imprecise” language models are smaller, speedier, and nearly as accurate

TinyStories: How Small Can Language Models Be and Still Speak Coherent English? (2023)

Stuff we figured out about AI in 2023

Ask HN: How are Markov chains so different from tiny LLMs?

Launch HN: Talc AI (YC S23) – Test Sets for AI

Ten Noteworthy AI Research Papers of 2023

Low responsiveness of ML models to critical or deteriorating health conditions

Phi-2: Self-Extend Boosts Performance, Extends Context to 8k Without Training

Textbooks are all you need

Consortium launched to build the largest open LLM

Mechanical Turk workers are using AI to automate being human

Do Machine Learning Models Memorize or Generalize?

Outperforming LLMs with less training data and smaller model sizes

The tide is shifting: 1.3B outperforms 7B Llama 2

Apple bans ChatGPT use by employees, report says

Some Neural Networks Learn Language Like Humans

Modern language models refute Chomsky’s approach to language

MonadGPT – What would have happened if ChatGPT was invented in the 17th century?

How Small Can Language Models Be and Still Speak Coherent English?

Lojban: Constructed language to eliminate ambiguity from communication

The Darwinian argument for worrying about AI

How to scale LLMs better with an alternative to transformers

Artificial General Intelligence – A gentle introduction

Superintelligence: The Idea That Eats Smart People

LLM Powered Autonomous Agents

Integrated Information Theory

Expanding Transformer size without losing function or starting from scratch

Disney Packed Big Emotion into a Little Robot

The 100k Whys of AI

One idea leads to another.