Eight Things to Know about Large Language Models
8 things experts know about LLMs (large language models) and you should know about them too.
► Sponsor: Prodigy 👉 https://prodi.gy/
Check out our #MachineLearning Quiz Questions: https://www.youtube.com/c/AICoffeeBreak/community
📜 Bowman, Samuel R. "Eight things to know about large language models." arXiv preprint arXiv:2304.00612 (2023). https://arxiv.org/abs/2304.00612
📺 Debate On AGI: Existential or Non-existential? (Connor Leahy, Joseph Jacks) [MLST LIVE] https://www.youtube.com/live/Mw7BHDboARY
📺 OpenAI suggests AI licenses (US Senate hearing on AI regulation w/ Sam Altman) https://youtu.be/I72_GJHzH3Y
☕ ChatGPT vs. Sparrow: https://youtu.be/SWwQ3k-DWyo
☕ Why does ChatGPT lie: https://youtu.be/XstVY5epRWs
☕ Language models vs. meaning: https://youtu.be/R80kc5cwEUw
☕ Can AI be moral? https://youtu.be/X_RKCTpuYRA
Thanks to our Patrons who support us in Tier 2, 3, 4: 🙏
Dres. Trost GbR, Siltax, Kshitij, Edvard Grødem, Vignesh Valliappan, Mutual Information
Outline:
00:00 Eight facts about Large Language Models
01:15 Prodigy (Sponsor)
02:15 LLMs complete sentences (like ChatGPT)
02:44 (1) LLMs get better with size
04:06 (2) LLMs surprise with emergent abilities
06:32 (3) LLMs seem to have "mental models"
07:45 (4) You cannot reliably steer LLMs
09:13 (5) Nobody knows why LLMs work
10:46 (6) They can outperform humans
11:35 (7) AI Alignment
11:41 (8) LLMs make for a great first impression
13:15 (9) They do not train for truth
▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀
🔥 Optionally, pay us a coffee to help with our Coffee Bean production! ☕
Patreon: https://www.patreon.com/AICoffeeBreak
Ko-fi: https://ko-fi.com/aicoffeebreak
▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀
🔗 Links:
AICoffeeBreakQuiz: https://www.youtube.com/c/AICoffeeBreak/community
Twitter: https://twitter.com/AICoffeeBreak
Reddit: https://www.reddit.com/r/AICoffeeBreak/
YouTube: https://www.youtube.com/AICoffeeBreak
#AICoffeeBreak #MsCoffeeBean #MachineLearning #AI #research
Video editing: Nils Trost
Видео Eight Things to Know about Large Language Models канала AI Coffee Break with Letitia
► Sponsor: Prodigy 👉 https://prodi.gy/
Check out our #MachineLearning Quiz Questions: https://www.youtube.com/c/AICoffeeBreak/community
📜 Bowman, Samuel R. "Eight things to know about large language models." arXiv preprint arXiv:2304.00612 (2023). https://arxiv.org/abs/2304.00612
📺 Debate On AGI: Existential or Non-existential? (Connor Leahy, Joseph Jacks) [MLST LIVE] https://www.youtube.com/live/Mw7BHDboARY
📺 OpenAI suggests AI licenses (US Senate hearing on AI regulation w/ Sam Altman) https://youtu.be/I72_GJHzH3Y
☕ ChatGPT vs. Sparrow: https://youtu.be/SWwQ3k-DWyo
☕ Why does ChatGPT lie: https://youtu.be/XstVY5epRWs
☕ Language models vs. meaning: https://youtu.be/R80kc5cwEUw
☕ Can AI be moral? https://youtu.be/X_RKCTpuYRA
Thanks to our Patrons who support us in Tier 2, 3, 4: 🙏
Dres. Trost GbR, Siltax, Kshitij, Edvard Grødem, Vignesh Valliappan, Mutual Information
Outline:
00:00 Eight facts about Large Language Models
01:15 Prodigy (Sponsor)
02:15 LLMs complete sentences (like ChatGPT)
02:44 (1) LLMs get better with size
04:06 (2) LLMs surprise with emergent abilities
06:32 (3) LLMs seem to have "mental models"
07:45 (4) You cannot reliably steer LLMs
09:13 (5) Nobody knows why LLMs work
10:46 (6) They can outperform humans
11:35 (7) AI Alignment
11:41 (8) LLMs make for a great first impression
13:15 (9) They do not train for truth
▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀
🔥 Optionally, pay us a coffee to help with our Coffee Bean production! ☕
Patreon: https://www.patreon.com/AICoffeeBreak
Ko-fi: https://ko-fi.com/aicoffeebreak
▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀
🔗 Links:
AICoffeeBreakQuiz: https://www.youtube.com/c/AICoffeeBreak/community
Twitter: https://twitter.com/AICoffeeBreak
Reddit: https://www.reddit.com/r/AICoffeeBreak/
YouTube: https://www.youtube.com/AICoffeeBreak
#AICoffeeBreak #MsCoffeeBean #MachineLearning #AI #research
Video editing: Nils Trost
Видео Eight Things to Know about Large Language Models канала AI Coffee Break with Letitia
Показать
Комментарии отсутствуют
Информация о видео
7 июня 2023 г. 16:19:58
00:14:46
Другие видео канала
What is LoRA? Low-Rank Adaptation for finetuning LLMs EXPLAINEDAre ChatBots their own death? | Training on Generated Data Makes Models Forget – Paper explainedThe first law on AI regulation | The EU AI ActSay that 3 times in a row. 😅Author Interviews, Poster Highlights, Summary of the ACL 2023 Toronto NLPChatGPT ist not an intelligent agent. It is a cultural technology. – Gopnik KeynoteDo LLMs understand? Jay Alammar's TLDR of Geoffrey Hinton ACL2023 Keynote[Own work] MM-SHAP to measure modality contributionsSpeaking about AI is hard, even for humans | AI Coffee Break BloopersMoral Self-Correction in Large Language Models | paper explainedAI beats us at another game: STRATEGO | DeepNash paper explainedWhy ChatGPT fails | Language Model Limitations EXPLAINED"Watermarking Language Models" paper and GPTZero EXPLAINED | How to detect text by ChatGPT?Training learned optimizers: VeLO paper EXPLAINEDChatGPT vs Sparrow - Battle of ChatbotsPaella: Text to image FASTER than diffusion models | Paella paper explainedGenerate long form video with Transformers | Phenaki from Google Brain explainedMovie Diffusion explained | Make-a-Video from MetaAI and Imagen Video from Google BrainBeyond neural scaling laws – Paper ExplainedHow does Stable Diffusion work? – Latent Diffusion Models EXPLAINED