Загрузка видео...

Не удалось загрузить видео

На главную

Generic LLMs answer medical questions confidently. They're also often wrong. We fine-tuned one on PubMed literature with gradients sdk. Now it agrees with the peer-reviewed answer. 3 lines of Python: pip install gradientsio

gradients_ai's profile picture

Gradients

2,411 subscribers

10,647 просмотров • 4 месяцев назад •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

Комментарии: 1

Фото профиля amiropensourcefan τ
amiropensourcefan τ4 месяцев назад

@const_reborn Seems #gradients #SN56 is waking up!

Похожие видео

🔥 We fine-tuned Gemma 3 using Unsloth AI on a custom instruction dataset scraped and built with . The model is now able to answer questions about OpenAI Agents SDK documentation. Check out the demo below👇
0:48

🔥 We fine-tuned Gemma 3 using Unsloth AI on a custom instruction dataset scraped and built with . The model is now able to answer questions about OpenAI Agents SDK documentation. Check out the demo below👇

Eric Ciarla (hiring)

28,794 просмотров • 1 год назад

One week left until the interview with the Duffer Brothers, where they'll answer all the big questions about the Stranger Things finale. David Harbour also sent a question for the Duffers... 👀 Stay tuned, follow me and turn on notifications so you don't miss anything!
0:29

One week left until the interview with the Duffer Brothers, where they'll answer all the big questions about the Stranger Things finale. David Harbour also sent a question for the Duffers... 👀 Stay tuned, follow me and turn on notifications so you don't miss anything!

Stranger Things Updates

29,224 просмотров • 4 месяцев назад

A New Zealand psychology journal has removed a peer-reviewed paper. Not for fraud. Not for error. But because it didn't align with the college's values. The author is a Māori clinical psychologist. Her paper was pulled to "protect Māori". The answer to a paper you disagree with is to argue back, not to take it down. Removing it doesn't refute it. It just tells every clinician that some questions are now off the table.
1:45

A New Zealand psychology journal has removed a peer-reviewed paper. Not for fraud. Not for error. But because it didn't align with the college's values. The author is a Māori clinical psychologist. Her paper was pulled to "protect Māori". The answer to a paper you disagree with is to argue back, not to take it down. Removing it doesn't refute it. It just tells every clinician that some questions are now off the table.

🗣 Free Speech Union ✊

34,645 просмотров • 3 месяцев назад

“We are very close together, on the pitch and also off the pitch. We’re just like family.” ❤️ Wataru Endo sat down with Florence and Grace to answer all of their questions 🙌 Standard Chartered Play On 🗣️ #Ad
2:00

“We are very close together, on the pitch and also off the pitch. We’re just like family.” ❤️ Wataru Endo sat down with Florence and Grace to answer all of their questions 🙌 Standard Chartered Play On 🗣️ #Ad

Liverpool FC

273,062 просмотров • 11 месяцев назад

🚨 BREAKING Scientists may have just cracked the code on Alzheimer's and Parkinson's — WITH ONE PILL. A drug called Buntanetap was just published in a peer-reviewed journal with Phase 3 data. Phase 3 means that we are on the verge of FDA approval. It targets the ROOT CAUSE of both diseases simultaneously. Here's what you need to know 🧵👇 #Alzheimers #Parkinsons
0:20

🚨 BREAKING Scientists may have just cracked the code on Alzheimer's and Parkinson's — WITH ONE PILL. A drug called Buntanetap was just published in a peer-reviewed journal with Phase 3 data. Phase 3 means that we are on the verge of FDA approval. It targets the ROOT CAUSE of both diseases simultaneously. Here's what you need to know 🧵👇 #Alzheimers #Parkinsons

Harshi Peiris, Ph.D.

504,674 просмотров • 3 месяцев назад

The problem with Rupert’s stock answer on the Iran War is he uses Britain’s military weakness as a bit of a cop out as to why we can’t help the US. Which suggests that if we were stronger he’d think it was a good idea to join the war. Which I’m afraid is simply the wrong answer.
1:36

The problem with Rupert’s stock answer on the Iran War is he uses Britain’s military weakness as a bit of a cop out as to why we can’t help the US. Which suggests that if we were stronger he’d think it was a good idea to join the war. Which I’m afraid is simply the wrong answer.

Wolf 🐺

14,226 просмотров • 5 месяцев назад

We skipped our own wedding and flew to SF. LLMs are the last piece of software that application developers still can’t easily build themselves. We’re letting every developer build their own LLMs in TypeScript, not Python. It’s Next.js + Vercel, but for building LLMs. In the last 7 days, users made 80k+ calls to models they fine-tuned. 3 devtool companies are running Arkor in their production pipelines to serve task-specific models with lower latency. Even users with PhDs in ML love the convenience of handling the entire training + deployment flow in Node.js. Meet Arkor. The best TypeScript framework for fine-tuning & deploying open models.
0:41

We skipped our own wedding and flew to SF. LLMs are the last piece of software that application developers still can’t easily build themselves. We’re letting every developer build their own LLMs in TypeScript, not Python. It’s Next.js + Vercel, but for building LLMs. In the last 7 days, users made 80k+ calls to models they fine-tuned. 3 devtool companies are running Arkor in their production pipelines to serve task-specific models with lower latency. Even users with PhDs in ML love the convenience of handling the entire training + deployment flow in Node.js. Meet Arkor. The best TypeScript framework for fine-tuning & deploying open models.

Hina

53,820 просмотров • 1 месяц назад

With the rise of 17+ games on Roblox focused on bars, I wondered how hard it would be to make a *basic* drink-pouring system in VR. The answer is <400 lines and about 3 hours of work gets the job done. Demo game: #RobloxDev
0:57

With the rise of 17+ games on Roblox focused on bars, I wondered how hard it would be to make a *basic* drink-pouring system in VR. The answer is <400 lines and about 3 hours of work gets the job done. Demo game: #RobloxDev

Nexus

29,077 просмотров • 3 лет назад

YOUR REACTION: Trump on ABC News debate on September 10: "We're thinking about it. They also want to change the rules. You know, the deal was we keep the same rules. Now, all of a sudden they want to make a change in the rules because she can't answer questions."
4:30

YOUR REACTION: Trump on ABC News debate on September 10: "We're thinking about it. They also want to change the rules. You know, the deal was we keep the same rules. Now, all of a sudden they want to make a change in the rules because she can't answer questions."

Simon Ateba

62,049 просмотров • 2 лет назад

Our Customer Support team gets thousands of questions a week. So we built a Custom Agent to help. It searches everything, responds with citations, and gets the team to the right answer fast. When it can't find one, it files a task so we can document the gap. And next time, it knows. Here's how we built it:
5:10

Our Customer Support team gets thousands of questions a week. So we built a Custom Agent to help. It searches everything, responds with citations, and gets the team to the right answer fast. When it can't find one, it files a task so we can document the gap. And next time, it knows. Here's how we built it:

Notion

16,506 просмотров • 6 месяцев назад

This is wild ChatGPT 6 is asked how many of the letter “E” are in the word seventeen. It can’t get the correct answer no matter how many times you ask. Even if you make it count out the letters, it still can’t get it right It will actually get an attitude and show frustration with you for trying to get the model to say the correct answer I ran the test multiple times and it keeps getting it wrong. When it does get it right, if you ask why it keeps getting it wrong it will then “correct” itself back to the wrong answer…. It even claims to be the smartest AI available According to Grok many AI models get 17 wrong. It has to do with LLMs. I asked Grok and it gets it right no problem It’s really wild that we are putting so much faith in AI when currently there are issues like this on such simple things
3:17

This is wild ChatGPT 6 is asked how many of the letter “E” are in the word seventeen. It can’t get the correct answer no matter how many times you ask. Even if you make it count out the letters, it still can’t get it right It will actually get an attitude and show frustration with you for trying to get the model to say the correct answer I ran the test multiple times and it keeps getting it wrong. When it does get it right, if you ask why it keeps getting it wrong it will then “correct” itself back to the wrong answer…. It even claims to be the smartest AI available According to Grok many AI models get 17 wrong. It has to do with LLMs. I asked Grok and it gets it right no problem It’s really wild that we are putting so much faith in AI when currently there are issues like this on such simple things

Wall Street Apes

57,198 просмотров • 13 дней назад

Can AI tell when a question is total nonsense, or does it just make up an answer? Peter Gostev (SF 24-28 August) tested 80 models with nonsense questions. Some pushed back. Others confidently invented fake metrics and kept going. All of them were ranked on the "BS Bench". One surprise: thinking harder made it worse. Watch the full deep dive on BS Bench on YouTube. Link in thread.
1:06

Can AI tell when a question is total nonsense, or does it just make up an answer? Peter Gostev (SF 24-28 August) tested 80 models with nonsense questions. Some pushed back. Others confidently invented fake metrics and kept going. All of them were ranked on the "BS Bench". One surprise: thinking harder made it worse. Watch the full deep dive on BS Bench on YouTube. Link in thread.

Arena.ai

10,943 просмотров • 6 месяцев назад

Bidet, Critters! To celebrate "The Legend of Vox Machina" Season 4, we had the #CriticalRole cast play with puppies and answer questions! They chat about their favorite voice acting roles, dice, love languages, and more. Watch it on BuzzFeed Celeb:
0:45

Bidet, Critters! To celebrate "The Legend of Vox Machina" Season 4, we had the #CriticalRole cast play with puppies and answer questions! They chat about their favorite voice acting roles, dice, love languages, and more. Watch it on BuzzFeed Celeb:

BuzzFeed Celeb

75,258 просмотров • 3 месяцев назад

Why should America support Israel? It’s a hot topic right now—and a deeply important one. As Christians, why should we stand with Israel and the Jewish people? What does the Bible say? I answer these questions—and more—in this video. Check it out!
3:35

Why should America support Israel? It’s a hot topic right now—and a deeply important one. As Christians, why should we stand with Israel and the Jewish people? What does the Bible say? I answer these questions—and more—in this video. Check it out!

Greg Laurie

53,208 просмотров • 1 год назад

Do sharks cuddle? That question has a definite answer, and it might not be the one you'd assume. #HammerheadSharksUpClose with Bertie Gregory is now streaming on Disney+ and Hulu, part of #Sharkfest on National Geographic.
0:43

Do sharks cuddle? That question has a definite answer, and it might not be the one you'd assume. #HammerheadSharksUpClose with Bertie Gregory is now streaming on Disney+ and Hulu, part of #Sharkfest on National Geographic.

National Geographic

93,122 просмотров • 2 месяцев назад

Whenever your friend, mom, colleague... ask for help with a task on Mac or iOS, the answer is Setapp Family. You + 3 of your favorite people, 240+ apps, one Setapp plan. Sign up now👉
0:40

Whenever your friend, mom, colleague... ask for help with a task on Mac or iOS, the answer is Setapp Family. You + 3 of your favorite people, 240+ apps, one Setapp plan. Sign up now👉

Setapp by MacPaw

71,589 просмотров • 3 лет назад

MacCallum: Last night I ChatGPT’d, where is the finance center of the world and it said New York City… Mamdani: That is one thing you have in common with Andrew Cuomo because he also uses ChatGPT to answer many questions including how to resolve the housing crisis in the city.
0:48

MacCallum: Last night I ChatGPT’d, where is the finance center of the world and it said New York City… Mamdani: That is one thing you have in common with Andrew Cuomo because he also uses ChatGPT to answer many questions including how to resolve the housing crisis in the city.

Acyn

779,217 просмотров • 11 месяцев назад

One of the biggest KOLs agrees with me on: $ASTRA AstraLabs Inc. Alex Becker 🍊🏆🥇 shared that $ASTRA is his favorite #AI token, that has what it takes to go on a huge rally. He also mentioned that he has often chats with the team 🤝 Breakout confirmed💯 $100M+ Marketcap⏳
0:25

One of the biggest KOLs agrees with me on: $ASTRA AstraLabs Inc. Alex Becker 🍊🏆🥇 shared that $ASTRA is his favorite #AI token, that has what it takes to go on a huge rally. He also mentioned that he has often chats with the team 🤝 Breakout confirmed💯 $100M+ Marketcap⏳

KingAce 🔥

11,262 просмотров • 1 год назад