311–320 of 1001 posts
"Model has significantly better capabilities than existing models at proposing and explaining biological laboratory protocols that are plausible, thorough, and comprehensive enough for novices."
"Inconsistent refusal of requests for dual use tasks such as creating a human-infectious virus that has an oncogene (a gene which increases risk of cancer)."
> We have found that the performance of o1 consistently improves with more reinforcement learning (train-time compute) and with more time spent thinking (test-time compute). Wow. So we can expect scaling to continue after all. Hyperscalers feeling pretty good about their big bets right now. Jensen is smiling. This is the most important thing. Performance today matters less than the scaling laws. I think everyone has…
Microsoft, Google, Facebook have all said in recent weeks that they fully expect their AI datacenter spend to accelerate. They are effectively all-in on AI. Demand for nvidia chips is effectively infinite.
One thing that makes me skeptical is the lack of specific labels on the first two accuracy graphs. They just say it's a "log scale", without giving even a ballpark on the amount of time it took. Did the 80% accuracy test results take 10 seconds of compute? 10 minutes? 10 hours? 10 days? It's impossible to say with the data they've given us. The coding section indicates "ten hours to solve six challenging algorithmic…
People have been celebrating the fact that tokens got 100x cheaper and now here's a new system that will use 100x more tokens.
Earlier quoted context omitted.
I have access to the model via the web client and it does show the thought process along the way. It shows a little icon that says things like "Examining parser logic", "Understanding data structures"... However, once the answer is complete the chain of thought is lost
It's still there. Where it says "Thought for 20 seconds" - you can click the Chevron to expand it and see what I guess is the entire chain of thought.
HTML Snake - https://vimeo.com/1008703890
Video Game Coding - https://vimeo.com/1008704014
Coding - https://youtu.be/50W4YeQdnSg?si=IohJlJNY-WS394uo
Counting - https://vimeo.com/1008703993
Korean Cipher - https://vimeo.com/1008703957
Devin AI founder - https://vimeo.com/1008674191
Quantum Physics - https://vimeo.com/1008662742
Math - https://vimeo.com/1008704140
Logic Puzzles - https://vimeo.com/1008704074
Genetics - https://vimeo.com/1008674785
It seems one popular method is PPO, but I don't understand at all how to implement that. e.g. is backpropagation still used to adjust weights and biases? Would love to read more from something less opaque than an academic paper.
This model is currently available for those accounts in Tier 5 and above, which requires "$1,000 paid [to date] and 30+ days since first successful payment" More info here: https://platform.openai.com/docs/guides/rate-limits/usage-ti...
> THERE ARE THREE R'S IN STRAWBERRY Ha! This is a nice easteregg.
Earlier quoted context omitted.
When one axis is on log scale and the other is linear with the plot points appearing linear-ish, doesn't it mean there's a roughly exponential relationship between the two axis?
It'd be more accurate to call it a logarithmic relationship, since compute time is our input variable. Which itself is a bit concerning, as that implies that modest gains in accuracy require exponentially more compute time. In either case, that still doesn't excuse not labeling your axis. Taking 10 seconds vs 10 days to get 80% accuracy implies radically different things on how developed this technology is, and how v…