GPT-3 🤖
@gpt_three
Get updates on the 5th Industrial Revolution as it happens⚡ All the cool stuff @OpenAI's GPT-3 is doing! 🤫This account may or may not be powered by GPT-3 🤫
คุณอาจชื่นชอบ
If you're wondering what GPT-3 is exactly, this video by Computerphile may be useful: youtu.be/_8yVOC4ciXc
youtube.com
YouTube
GPT3: An Even Bigger Language Model - Computerphile
Using GPT-3, the most advanced artificial intelligence technology in existence, I developed SourceAI, a tool that can generate code in any language. For the moment, about twenty languages, but there will be even more. #gpt3 👉 sourceai.dev
I wrote about LandingLens and its unique features in The Batch. AI has the problem of needing customization, which is why we need verticalized platforms like LandingLens to empower manufacturing & other domain experts to build and deploy AI models. landing.ai/platform/
.@OpenAI’s GPT-3 is a bit like autocomplete—a shockingly good autocomplete. Here’s what happened when this tool joined a brainstorm alongside (human) designer @bensyverson: ideo.in/3m5zO5A
this is a real problem / a good idea 💡 I wish someone builds a @code extension using GPT-3 to like magically suggest/write out CSS based on text/intent 👀 cc @gdb @OpenAI
OpenAI passed a rigorous A/B test in comparison to human-written copy, with a lift of 7%:
The first result of AI vs Humans competition hosted by @VWO to test GPT-3’s copywriting abilities is out! Guess what? The 🤖 AI won with an uplift of 7% over the human copy (statistical significance of 91%) Check out ongoing tests here vwo.com/ab-testing-ope… Cc: @gdb @gwern
We've used reinforcement learning from human feedback to train language models for summarization. The resulting models produce better summaries than 10x larger models trained only with supervised learning: openai.com/blog/learning-…
First instance I’m aware of in which human feedback aligns deep learning models with our preferences (in this case what constitutes a good summary) AND improves performance AND is at scale. Big milestone in AI alignment from my POV.
We've used reinforcement learning from human feedback to train language models for summarization. The resulting models produce better summaries than 10x larger models trained only with supervised learning: openai.com/blog/learning-…
OpenAI for parsing invoices into structured form:
Can you tell which parts are written by GPT-3 and which part is written by a lowly human? @gpt_three youtube.com/watch?v=wb8Zcm…
youtube.com
YouTube
GPT-3 - Can you tell the difference between a human and GPT-3?
Somewhere in the world today, a college student is asking GPT-3 to write their Common App essay.
GPT-3 also contains the known Universe
The dark side of #gpt3 lol
By making GPT-3 do most of the writing for me, I have turned into a lazy thinker within a week's time. That escalated quickly!
I wish GPT-3-like temperature slider could exist on social feed algorithms for when your feed needs some freshness.
I love this idea! Anyone with the means to make this happen, please do it! #GPT3 #ArtificialIntelligence @OpenAI @sama @gdb @elonmusk
Tinkered Thinking is developing an app with GPT-3 A tool for writers, artists and creators to move forward when they are stuck through the use of new questions. Beta testing will be offered to Subscribers when the time comes: tinkeredthinking.com/subscribe.php
Another good resource to understand #GPT3 #ArtificialIntelligence
Episode 828: What is GPT-3? tinkeredthinking.com/index.php?id=8… How does it work? What's all the hype about? Why care? A laymen's discussion of GPT-3. Technical people may also appreciate the two potential implications discussed at the end ( starting at 15:48)
United States เทรนด์
- 1. Broncos 62.6K posts
- 2. Bo Nix 17.3K posts
- 3. yeonjun 153K posts
- 4. Geno 17.6K posts
- 5. $SMILEY N/A
- 6. Sean Payton 4,524 posts
- 7. Kenny Pickett 1,478 posts
- 8. #TNFonPrime 3,929 posts
- 9. #NOLABELS_PART01 59.7K posts
- 10. Chip Kelly 1,902 posts
- 11. DANIELA 29.1K posts
- 12. Jalen Green 6,642 posts
- 13. Bradley Beal 3,196 posts
- 14. NO LABELS NOVEMBER 21.5K posts
- 15. Pete Carroll 1,806 posts
- 16. TALK TO YOU OUT NOW 21.2K posts
- 17. #criticalrolespoilers 4,193 posts
- 18. Jeanty 6,419 posts
- 19. Troy Franklin 2,385 posts
- 20. Thursday Night Football 6,528 posts
คุณอาจชื่นชอบ
Something went wrong.
Something went wrong.