Published Aug 8, 2024

Interviewing Ross Taylor on LLM reasoning, Llama fine-tuning, Galactica, agents

Ross Taylor, former LLM lead at Meta AI, discusses his journey from founding Papers with Code to developing innovative language models like Galactica, sharing insights on fine-tuning techniques, reasoning in LLMs, and the future of AI agents.

Episode Highlights

Topics covered

Episode Highlights

Related Episodes

Interviewing Sebastian Raschka on the state of open LLMs, Llama 3.1, and AI education
Answers 383 questions
Interviewing Louis Castricato of Synth Labs and Eleuther AI on RLHF, Gemini Drama, DPO, founding Carper AI, preference data, reward models, and everything in between
Answers 383 questions
Llama 3: Scaling open LLMs to AGI
Answers 383 questions
Open Language Models (OLMos) and the LLM landscape
Answers 383 questions
Interviewing Tri Dao and Michael Poli of Together AI on the future of LLM architectures
Answers 383 questions
A recipe for frontier model post-training
Answers 383 questions
Interviewing Riley Goodside on the science of prompting
Answers 383 questions
GPT-4o-mini changed ChatBotArena
Answers 383 questions
Local LLMs, some facts some fiction
Answers 383 questions
Llama 3.1 405b, Meta's AI strategy, and the new open frontier model ecosystem
Answers 383 questions
RLHF: A thin line between useful and lobotomized
Answers 383 questions
Google ships it: Gemma open LLMs and Gemini backlash
Answers 383 questions
Big Tech's LLM evals are just marketing
Answers 383 questions
Llama 3.2 Vision and Molmo: Foundations for the multimodal open-source ecosystem
Answers 383 questions
A realistic path to robotic foundation models
Answers 383 questions

Dexa/Interconnects Audio

Interviewing Ross Taylor on LLM reasoning, Llama fine-tuning, Galactica, agents

Topics covered

Popular Clips

Future of AI Agents

Successful Model Teams

Language Models & Discovery

Model Fine-Tuning Insights

Model Training Insights

Model Evaluation Insights

Risk and Innovation

Closed Models Evolution

ML Startup Journey

Reinforcement Learning Insights

Market Saturation Insights

Methodology Matters

Model Reliability Insights

Future of Language Models

LLM Trade-offs

Episode Highlights

Startup Journey

Language Model Innovations

Future of AI

Related Episodes

Interviewing Sebastian Raschka on the state of open LLMs, Llama 3.1, and AI education

Interviewing Louis Castricato of Synth Labs and Eleuther AI on RLHF, Gemini Drama, DPO, founding Carper AI, preference data, reward models, and everything in between

Llama 3: Scaling open LLMs to AGI

Open Language Models (OLMos) and the LLM landscape

Interviewing Tri Dao and Michael Poli of Together AI on the future of LLM architectures

A recipe for frontier model post-training

Interviewing Riley Goodside on the science of prompting

GPT-4o-mini changed ChatBotArena

Local LLMs, some facts some fiction

Llama 3.1 405b, Meta's AI strategy, and the new open frontier model ecosystem

RLHF: A thin line between useful and lobotomized

Google ships it: Gemma open LLMs and Gemini backlash

Big Tech's LLM evals are just marketing

Llama 3.2 Vision and Molmo: Foundations for the multimodal open-source ecosystem

A realistic path to robotic foundation models

Interviewing Ross Taylor on LLM reasoning, Llama fine-tuning, Galactica, agents

Topics covered

Popular Clips

Episode Highlights

Startup JourneyRoss Taylor, a former LLM lead at Meta AI, shares his journey from founding Papers with Code to leading groundbreaking projects like Galactica. He discusses the challenges and milestones in developing language models and their potential to revolutionize scientific research.

Startup Journey

Language Model InnovationsRoss Taylor delves into the intricacies of the Galactica project, reasoning in language models, and fine-tuning techniques. He shares valuable insights on the challenges and advancements in these areas, offering a comprehensive look at the future of LLMs.

Language Model Innovations

Future of AIRoss Taylor explores the future of AI agents and their potential to revolutionize various fields. He discusses creative applications and the challenges in developing these advanced systems.

Future of AI

Related Episodes