Published Apr 14, 2023

670: LLaMA: GPT-3 performance, 10x smaller — with Jon Krohn (@JonKrohnLearns)

Explore Meta AI's revolutionary LLaMa model, shedding light on how it matches GPT-3's capabilities with a fraction of the size and carbon footprint, thanks to Chinchilla scaling laws. Dive into its commercial potential and the implications of its open-source dataset training with Jon Krohn.

Episode Highlights

Topics covered

Popular Clips

Episode Highlights

Related Episodes

824: Llama 3.2: Open-Source Edge and Multimodal LLMs — with Jon Krohn (@JonKrohnLearns)
Answers 383 questions
666: GPT-4 — with Jon Krohn (@JonKrohnLearns)
Answers 383 questions
806: Llama 3.1 405B: The First Open-Source Frontier LLM — with Jon Krohn (@JonKrohnLearns)
Answers 383 questions
712: Code Llama — with Jon Krohn (@JonKrohnLearns)
Answers 383 questions
672: Open-source "ChatGPT": Alpaca, Vicuña, GPT4All-J, and Dolly 2.0 — with @JonKrohnLearns
Answers 383 questions
702: Llama 2 — It's Time to Upgrade your Open-Source LLM — with Jon Krohn (@JonKrohnLearns)
Answers 383 questions
678: StableLM: Open-source "ChatGPT"-like LLMs you can fit on one GPU — with @JonKrohnLearns
Answers 383 questions
704: Jon’s “Generative A.I. with LLMs” Hands-on Training — with Jon Krohn (@JonKrohnLearns)
Answers 383 questions
SDS 568: PaLM: Google's Breakthrough Natural Language Model — with Jon Krohn
Answers 383 questions
768: Is Claude 3 Better than GPT-4? — with Jon Krohn (@JonKrohnLearns)
Answers 383 questions
706: Large Language Model Leaderboards and Benchmarks — with Caterina Constantinescu
Answers 383 questions
676: The Chinchilla Scaling Laws — with Jon Krohn (@JonKrohnLearns)
Answers 383 questions
728: Use Contrastive Search to get Human-Quality LLM Outputs — with Jon Krohn (@JonKrohnLearns)
Answers 383 questions
713: Llama 2, Toolformer and BLOOM: Open-Source LLMs — with Meta's Dr. Thomas Scialom
Answers 383 questions
772: In Case You Missed It in March 2024 — with Jon Krohn (@JonKrohnLearns)
Answers 383 questions

Dexa/Super Data Science: ML & AI Podcast with Jon Krohn

670: LLaMA: GPT-3 performance, 10x smaller — with Jon Krohn (@JonKrohnLearns)

Topics covered

Popular Clips

Llama Model Insights

Llama Model Insights

Llama Model Insights

Episode Highlights

LLaMa Model

Commercial Prospects

Chinchilla Laws

Related Episodes

824: Llama 3.2: Open-Source Edge and Multimodal LLMs — with Jon Krohn (@JonKrohnLearns)

666: GPT-4 — with Jon Krohn (@JonKrohnLearns)

806: Llama 3.1 405B: The First Open-Source Frontier LLM — with Jon Krohn (@JonKrohnLearns)

712: Code Llama — with Jon Krohn (@JonKrohnLearns)

672: Open-source "ChatGPT": Alpaca, Vicuña, GPT4All-J, and Dolly 2.0 — with @JonKrohnLearns

702: Llama 2 — It's Time to Upgrade your Open-Source LLM — with Jon Krohn (@JonKrohnLearns)

678: StableLM: Open-source "ChatGPT"-like LLMs you can fit on one GPU — with @JonKrohnLearns

704: Jon’s “Generative A.I. with LLMs” Hands-on Training — with Jon Krohn (@JonKrohnLearns)

SDS 568: PaLM: Google's Breakthrough Natural Language Model — with Jon Krohn

768: Is Claude 3 Better than GPT-4? — with Jon Krohn (@JonKrohnLearns)

706: Large Language Model Leaderboards and Benchmarks — with Caterina Constantinescu

676: The Chinchilla Scaling Laws — with Jon Krohn (@JonKrohnLearns)

728: Use Contrastive Search to get Human-Quality LLM Outputs — with Jon Krohn (@JonKrohnLearns)

713: Llama 2, Toolformer and BLOOM: Open-Source LLMs — with Meta's Dr. Thomas Scialom

772: In Case You Missed It in March 2024 — with Jon Krohn (@JonKrohnLearns)

670: LLaMA: GPT-3 performance, 10x smaller — with Jon Krohn (@JonKrohnLearns)

Topics covered

Popular Clips

Episode Highlights

LLaMa ModelJon Krohn explores Meta AI's LLaMa, a natural language model that rivals GPT-3 in performance while being significantly smaller. Utilizing the Chinchilla scaling laws, LLaMa demonstrates how extended training on smaller models can achieve remarkable results.

LLaMa Model

Commercial ProspectsLLaMa models, developed by Meta AI, offer GPT-3 level performance with a significantly smaller size and carbon footprint. These models are ideal for research applications but come with commercial use restrictions.

Commercial Prospects

Chinchilla LawsLLaMa, Meta AI's latest language model, showcases impressive performance by training smaller models for longer durations. This innovative approach, based on Chinchilla scaling laws, allows LLaMa to rival larger models like GPT-3 while utilizing open-source datasets for training.

Chinchilla Laws

Related Episodes