GPT-3

OpenAI Built an AI to Hack Its Own Code—Here’s What It Found

a16z Deep Dives

Earlier frontier model; limited for real security automation tasks before GPT-4.

Outlasting Noam Shazeer, Crowdsourcing Chai AI w/ 1.4m DAU — with William Beauchamp, Chai Research

Latent Space

A large language model mentioned as being considered too powerful to release by OpenAI during its early development.

Neal Stephenson: Sci-Fi, Space, Aliens, AI, VR & the Future of Humanity | Lex Fridman Podcast #240

Lex Fridman

A large language model, described as self-supervised and capable of generating text and conversing with humans after processing vast amounts of human-created content.

The 10,000x Yolo Researcher Metagame — with Yi Tay of Reka

Latent Space

GPT-3's release marked the emergence of few-shot and in-context learning, changing the paradigm of large language model research.

This World Does Not Exist — Joscha Bach, Karan Malhotra, Rob Haisfield (WorldSim, WebSim, Liquid AI)

Latent Space

Discussed as a powerful model whose capabilities were unlocked by World Sim and Web Sim, showing its potential beyond chat interfaces.

High Agency Pydantic over VC Backed Frameworks — with Jason Liu of Instructor

Latent Space

An early language model that Jason Lou was skeptical of, but later acknowledged its capabilities after ChatGPT's release.

Can We Contain Artificial Intelligence?: A Conversation with Mustafa Suleyman (Episode #332)

Sam Harris

A large language model launched in 2020, with significantly fewer parameters now available in open-source versions.

[Paper Club] Weight Streaming on Wafer-Scale Clusters (w/ Sarah Chieng of Cerebras)

Latent Space

A large language model mentioned in the context of weight sparsity research, where Cerebras demonstrated creating a sparse representation of GPT-3 without losing accuracy.

Supervise the Process of AI Research — with Jungwon Byun and Andreas Stuhlmüller of Elicit

Latent Space

A significant language model release that prompted Elicit to shift focus towards building a more general research assistant.

Agents @ Work: Dust.tt — with Stanislas Polu

Latent Space

A key product from OpenAI that received significant compute resources.

Why Google failed to make GPT-3 -- with David Luan of Adept

Latent Space

A subsequent large language model from OpenAI, its scaling up presented a challenge and stress for research teams at Google due to resource competition.

A Comprehensive Overview of Large Language Models - Latent Space Paper Club

Latent Space

A large language model by OpenAI, notable for its powerful zero-shot learning abilities.

Building the Silicon Brain - Drew Houston of Dropbox

Latent Space

The model that, along with ChatGPT, significantly advanced the capabilities of LLMs and became accessible via API.

Let's build GPT: from scratch, in code, spelled out.

Andrej Karpathy

Production AI Engineering starts with Evals

Latent Space

An advanced language model by OpenAI, which 'totally blew the speaker's mind' with its ability to extract information from unstructured text, even without visual signals, surpassing LayoutLM.

The Ultimate Guide to Prompting - with Sander Schulhoff from LearnPrompting.org

Latent Space

An earlier large language model that Sander Schulhoff used for a translation task, marking his first introduction to prompting.

[Paper Club] 🍓 On Reasoning: Q-STaR and Friends!

Latent Space

Mentioned in the context of understanding reasoning, with the STAR paper providing a way to train models for this.

Is finetuning GPT4o worth it?

Latent Space

An early language model from OpenAI that inspired the founders of Cosign to explore AI for coding tasks.

Douglas Lenat: Cyc and the Quest to Solve Common Sense Reasoning in AI | Lex Fridman Podcast #221

Lex Fridman

A language model cited as an example of systems with 'wacky brittleness' that fail at common sense reasoning despite impressive statistical capabilities.

The Origin and Future of RLHF: the secret ingredient for ChatGPT - with Nathan Lambert

Latent Space

An earlier OpenAI language model.

What Do We Know About Our Minds?: A Conversation with Paul Bloom (Episode #317)

Sam Harris

An earlier version of the GPT language model, mentioned as being used by Sam Harris to generate fabricated quotes for an article, illustrating issues with AI hallucination.

The AI-First Graphics Editor - with Suhail Doshi of Playground AI

Latent Space

An early language model from OpenAI, which had a playground interface and was considered for address bar prediction by Mighty.

Jim Keller: The Future of Computing, AI, Life, and Consciousness | Lex Fridman Podcast #162

Lex Fridman

A language model by OpenAI, remarkable for demonstrating unsupervised learning capabilities with essentially infinite data.

Jaron Lanier: Virtual Reality, Social Media & the Future of Humans and AI | Lex Fridman Podcast #218

Lex Fridman

A large language model that Lanier's office is funding, which he notes still relies on 'statistical emergent pseudosemantics' and lacks deep representation.