Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

An analysis of DeepSeek's R1-Zero and R1

arcprize.org

Post reply on HN
Print view View original

191–200 of 280 posts

Re: An analysis of DeepSeek's R1-Zero and R1

#191

Post by notpushkin »

blibble wrote: post #106
BorisMelnik wrote: post #60

Earlier quoted context omitted.

I am not in this space, question: are there "bad actors" that are known to feed AI models with poisonous information?

yes, example: me I more often than not use the thumbs up on bad Google AI answers (but not always! can't find me that easily!)

I deliberately pick wrong answers in reCAPTCHA sometimes. I’ve found out that the audio version accepts basically any string slightly resembling the audio, so that’s the easiest way. (Images on the other hand punish you pretty hard at times – even if you solve it correctly!)

Re: An analysis of DeepSeek's R1-Zero and R1

#192

Post by vidarh »

boroboro4 wrote: post #154
vidarh wrote: post #124

Earlier quoted context omitted.

> You can call up Nvidia and order $10B worth of GPUs and have them delivered the next week Nvidia sold $14.5 billion of datacenter hardware in the third quarter of their fiscal 2024 and that led to severe supply constraints, with estimate lead times for H100's up to 52 weeks some places, so no you can't, as that $14.5 billion was clearly capped by their ability to supply, not demand. You're right, though, that Groq…

Groq chips have 230 mb of sram memory. Good luck running 670B model on those chips, even without supply constraints.

Nvidia's H100 has 80GB. As long the interconnect is fast enough you don't need everything to fit on one.

Re: An analysis of DeepSeek's R1-Zero and R1

#193

Post by vidarh »

boroboro4 wrote: post #153
onlyrealcuzzo wrote: post #127

Earlier quoted context omitted.

> OpenAI, Meta, AWS, AMD, and others have long attempted to eliminate the Nvidia tax, yet failed. Gemini / Google runs and trains on TPUs. You have no incentive to infer on AMD if you need to buy a massive Nvidia cluster to train.

Meta trains on Nvidia and infers on AMD. There is incentive if your inference costs are high.

Meta also has a second generation of their own AI accelerator chips designed.

Re: An analysis of DeepSeek's R1-Zero and R1

#194

Post by pigeons »

Uehreka wrote: post #148
whilenot-dev wrote: post #114

Earlier quoted context omitted.

> As long as the vast majority of their data is good (and it will be) So expert answers are out of scope? Nice, looking forward to those quality data!

If you want to pick apart my hastily concocted examples, well, have fun I guess. My overall point is that ensuring data quality is something OpenAI is probably very good at. They likely have many clever techniques, some of which we could guess at, some of which would surprise us, all of which they’ve validated through extensive testing including with adversarial data. If people want to keep playing pretend that their…

I'm interested in why you think OpenAI is probably very good at ensuring data quality. Also interested if you are trying to troll the resistance into revealing their working techniques.

Re: An analysis of DeepSeek's R1-Zero and R1

#195

Post by godelski »

spyckie2 wrote: post #22

> But now with reasoning systems and verifiers, we can create brand new legitimate data to train on. This can either be done offline where the developer pays to create the data or at inference time where the end user pays! > This is a fascinating shift in economics and suggests there could be a runaway power concentrating moment for AI system developers who have the largest number of paying customers. Those customers…

  > I highly doubt you are getting novel, high quality data.
That's not the point. The point is you reject low quality data, aka noise

Re: An analysis of DeepSeek's R1-Zero and R1

#196

Post by vanviegen »

nemomarx wrote: post #162
genewitch wrote: post #89

Earlier quoted context omitted.

the date is in the "system prompt", so the cron job that updates the prompts to the current date may be in a different time zone than you. 7f5dbb71f54322f271c4d3fc3aaa4d3282a1af5541d82b2cbc5aa10c1420b6bc

why can't they feed in user data like time zone and locale?

They're not actually processing the entire system prompt (which is rather long) on every query, but continuing from a model state saved after processing the system prompt once.

That makes it a bit harder, but still, spitting out the wrong date just seems like a plain old time-zone bug.

Re: An analysis of DeepSeek's R1-Zero and R1

#198

Post by energy123 »

polishdude20 wrote: post #6

I predict that the future of LLM's when it comes to coding and software creation is in "custom individually tailored apps". Imagine telling an AI agent what app you want, the requirements and all that and it just builds everything needed from backend to frontend, asks for your input on how things should work, clarifying questions etc. It tests the software by compiling and running it reading errors and failed tests a…

A little further out from that could be the LLM acting as the runtime environment. No code. It's just data in (user inputs etc) -> GUI out.

Re: An analysis of DeepSeek's R1-Zero and R1

#199

Post by baq »

halfadot wrote: post #171
hammock wrote: post #54

Earlier quoted context omitted.

The AI models to begin with assume that a significant majority of the training material is honest/in good faith. So that is not new?

AI models don't assume anything. AI models are just statistical tools. Their data is prepared by humans, who aren't morons. What is it with these super-ignorant AI critiques popping up everywhere?

There’s so much data required for training it’d be surprising humans look at even a small subset of it at all. They need different statistical tools to clean it up. That’s where attacks will be concentrated, naturally, and this is why synthetic data will overtake real human data, just after ‘there isn’t enough data even if it’s too much already’.

Re: An analysis of DeepSeek's R1-Zero and R1

#200

Post by buyucu »

panabee wrote: post #135
buyucu wrote: post #132

Earlier quoted context omitted.

My company recently switched from A100s to MI300s. I can confidently say that in my line of work, there is no CUDA moat. Onboarding took about month, but afterwards everything was fine.

Alternatives exist, especially for mature and simple models. The point isn't that Nvidia has 100% market share, but rather that they command the most lucrative segment and none of these big spenders have found a way to quit their Nvidia addiction, despite concerted efforts to do so. For instance, we experimented with AWS Inferentia briefly, but the value prop wasn't sufficient even for ~2022 computer vision models. T…

llm inference is fine on rocm. llama.cpp and vllm both have very good rocm support.

llm training is also mostly fine. I have not encountered any issues yet.

most of the cuda moat comes from people who are repeating what they heard 5-10 years ago.

Post reply on HN