Quick new post: Auto-grading decade-old Hacker News discussions with hindsight I took all the 930 frontpage Hacker News article+discussion of December 2015 and asked the GPT 5.1 Thinking API to do an in-hindsight analysis to identify the most/least prescient comments. This took
@karpathy
-
Clarification on AI Prompt Techniques and Expert Positioning Methods
By
–
A good chunk of people misunderstood this tweet btw, which is my bad. I am not suggesting people use the old style promoting techniques of “you are an expert swift programmer” or etc. it’s ok.
-
Bug Fix in NanoGPT SpellingBee Task Evaluation
By
–
ty to ericsilberstein1 on github for spotting the bug. https://
github.com/karpathy/nanoc
hat/pull/306
…
(it's not a big bug and only comes up in the SpellingBee synthetic task evaluation but still). -
Python Random Seed Bug: Identical Streams from Different Seeds
By
–
In today's episode of programming horror… In the Python docs of random.seed() def, we're told
"If a is an int, it is used directly." [1] But if you seed with 3 or -3, you actually get the exact same rng object, producing the same streams. (TIL). In nanochat I was using the -
Nested Simulation Efficiency in AI Model Computing
By
–
I could certainly imagine that "nesting" the simulation might be too "effortful" for the model, compute or data density wise. My results with it are not too bad so imo it's at least worth people try / experiment with / think about. For example it might be useful to read multiple
-
Engineering Emergent Personalities in LLM Reward Optimization
By
–
There is definitely work going into engineering the "you" simulation – the personality that gets all the rewards in verifiable problems, or all the upvotes from users/judge LLMs, or mimics the responses of SFT, and there is an emergent composite personality from that. My point is
-
LLMs as Simulators: Rethinking Prompt Engineering Approaches
By
–
Don't think of LLMs as entities but as simulators. For example, when exploring a topic, don't ask: "What do you think about xyz"? There is no "you". Next time try: "What would be a good group of people to explore xyz? What would they say?" The LLM can channel/simulate many
-
Edutainment: One Weird Trick to Guilt-Free Entertainment Consumption
By
–
Edutainment. This one weird trick to consume entertainment and feel good about it.
-
Turtle Graphics: Classic Programming Learning Tool Rediscovered
By
–
So cool! Distinctly reminded of turtle graphics (a popular way to learn programming)
-
AI Image Analysis Limitations Training Data Freshness
By
–
One more comment is that giving this image to an AI and asking about it is not sufficient to show the diff because it's all over the training data by now. You'd have to use a new, very recent image, taken yesterday or something. But it doesn't super matter, even if it didn't work