Published Nov 3, 2023
728: Use Contrastive Search to get Human-Quality LLM Outputs — with Jon Krohn (@JonKrohnLearns)
Jon Krohn delves into the transformative power of contrastive search technology in enhancing large language models' text generation, spotlighting its superiority over traditional methods like greedy and beam search, and analyzing advanced sampling techniques such as top k and nuclear sampling for more coherent outputs.
Topics covered
Popular Clips
Episode Highlights
Related Episodes
772: In Case You Missed It in March 2024 — with Jon Krohn (@JonKrohnLearns)
Answers 383 questionsSDS 464: A.I. vs Machine Learning vs Deep Learning — with Jon Krohn
Answers 383 questions670: LLaMA: GPT-3 performance, 10x smaller — with Jon Krohn (@JonKrohnLearns)
Answers 383 questions720: OpenAI’s DALL-E 3, Image Chat and Web Search — with Jon Krohn (@JonKrohnLearns)
Answers 383 questions684: Get More Language Context out of your LLM — with Jon Krohn (@JonKrohnLearns)
Answers 383 questions
808: In Case You Missed It in July 2024 — with Jon Krohn (@JonKrohnLearns)
Answers 383 questions666: GPT-4 — with Jon Krohn (@JonKrohnLearns)
Answers 383 questions750: How AI is Transforming Science — with Jon Krohn (@JonKrohnLearns)
Answers 383 questionsSDS 568: PaLM: Google's Breakthrough Natural Language Model — with Jon Krohn
Answers 383 questions740: Q*: OpenAI's Rumored AGI Breakthrough — with @JonKrohnLearns
Answers 383 questions