harb retweeted
The day you stop challenging yourself is the day you stop evolving. Surround yourself with people who challenge you.
harb retweeted
men will literally walk out on anything in 30 seconds flat if they feel the heat around the corner instead of going to therapy
harb retweeted
I have a major result to share.
I’ve found conclusive forensic evidence that the top 4 Chinese AI labs—DeepSeek, Moonshot, MiniMax and Ant Group—have built their newest models on NVIDIA’s most advanced Blackwell chips, which are barred from export to China.
This sounds insane, but I discovered that almost every AI model carries a hidden signature of the GPU that trained it. When a model is first set up, the starting values of its weights are laid out in a pattern that depends on the exact GPU. Each combination of GPU and seed number produces a unique pattern—a sort of fingerprint of the hardware used to create the model. Traces of that pattern survive training into the published model.
Again, this evidence isn’t based on any insider information or leaking. It’s based on a forensic analysis of the model weights released by the companies themselves.
I searched more than six trillion combinations of chip layout and seed number and matched them to open-weights models. In doing so, I found conclusive evidence that six Chinese AI models—including DeepSeek-V4-Pro, Kimi K3, and MiniMax-M3—were initialized on NVIDIA Blackwell systems. Four of the models match Grace Blackwell, the GB200 and GB300 parts in NVIDIA’s flagship liquid-cooled rack-scale systems.
I then verified these results on real hardware. Running unmodified PyTorch with the seeds I’d recovered, a Blackwell B300 reproduced DeepSeek-V4-Flash’s untouched starting weights bit for bit, and a rented GB200 reproduced Kimi K3’s to a correlation of 0.999995. A Hopper H100 reproduced none of DeepSeek-V4-Flash’s, and changing the seed by one broke the match. (DeepSeek’s seed is 953049583 by the way.) The method also was verified to match 24 known open-weights models to the hardware that developers said they used.
It is exceedingly unlikely that these results are a coincidence. Across six trillion combinations, the best accidental match should reach only about 8 sigma (i.e., eight standard deviations above what random chance would produce). The real matches range from 49 to 1,385 sigma, and even the weakest single test hits 20.
These were data-center-scale training runs. With some analysis of release schedules and model details, I estimate that the largest, Kimi K3 and DeepSeek-V4-Pro, used 2,000 to 3,000 Blackwell GPUs each. Clusters of this size cost on the order of $150 to $200 million each to buy, and across these labs alone would add up to more than half a billion dollars of export-controlled GPUs, not counting experimental compute or otherwise. It’s quite likely that we see multi-billion dollar expenditure on export-controlled GPUs.
I am highly confident in this result, and it has been independently verified by senior researchers at OpenAI and Anthropic. You can read the full report at unemploymentlabs.com/checkpo….
There is much more I’ve identified about these models through other forms of checkpoint forensics like this, and will have more to share soon.
One day, in the middle of a training block, I decided to count calories because my recovery was poor and my energy was low.
Turns out, I should have been eating way more. Tracking food is now is a daily habit
Runners will track every mile, analyze their heart rate, obsess over their training zones, and spend hundreds of dollars on running shoes.
Then you ask them what they eat in a typical day and they genuinely have no idea.
One of the most surprising things I’ve noticed coaching runners.
4
215
great video, relevant across many domains.
If you don't know what you're doing, you should be planting seeds everywhere. Doing a bit of everything has more expected value than committing to the wrong thing
I'm not a big creator, but I grew 33k followers across Youtube, Linkedin, Twitter and Instagram.
If you're a small creator, this is content 101:
58
harb retweeted
# on shortification of "learning"
There are a lot of videos on YouTube/TikTok etc. that give the appearance of education, but if you look closely they are really just entertainment. This is very convenient for everyone involved : the people watching enjoy thinking they are learning (but actually they are just having fun). The people creating this content also enjoy it because fun has a much larger audience, fame and revenue. But as far as learning goes, this is a trap. This content is an epsilon away from watching the Bachelorette. It's like snacking on those "Garden Veggie Straws", which feel like you're eating healthy vegetables until you look at the ingredients.
Learning is not supposed to be fun. It doesn't have to be actively not fun either, but the primary feeling should be that of effort. It should look a lot less like that "10 minute full body" workout from your local digital media creator and a lot more like a serious session at the gym. You want the mental equivalent of sweating. It's not that the quickie doesn't do anything, it's just that it is wildly suboptimal if you actually care to learn.
I find it helpful to explicitly declare your intent up front as a sharp, binary variable in your mind. If you are consuming content: are you trying to be entertained or are you trying to learn? And if you are creating content: are you trying to entertain or are you trying to teach? You'll go down a different path in each case. Attempts to seek the stuff in between actually clamp to zero.
So for those who actually want to learn. Unless you are trying to learn something narrow and specific, close those tabs with quick blog posts. Close those tabs of "Learn XYZ in 10 minutes". Consider the opportunity cost of snacking and seek the meal - the textbooks, docs, papers, manuals, longform. Allocate a 4 hour window. Don't just read, take notes, re-read, re-phrase, process, manipulate, learn.
And for those actually trying to educate, please consider writing/recording longform, designed for someone to get "sweaty", especially in today's era of quantity over quality. Give someone a real workout. This is what I aspire to in my own educational work too. My audience will decrease. The ones that remain might not even like it. But at least we'll learn something.
the typical running periodization is:
base + strides > sub-threshold (high-end aerobic) > vo2max > race specific
But you can honestly stop at sub-threshold and just stay there. You'll make great progress
Why I like it:
- lower injury risk
- can stack a ton of quality volume
you will see massive aerobic gains if you run a weekly interval session for medium-length reps 1-2 steps back from threshold effort
800m, 1000m, 1200m, 1500m, mile, 2000m, and 2 mile reps should be your best friend
run these at a comfy hard / fun effort
these intervals should be done at an effort that’s HMP or just a little quicker
if more runners stacked 16 weeks of sessions with this type of interval and intensity they will have a good race from 5k to 100 miles
i am a huge proponent in runners training at sub-threshold
there is more pieces to the pie including daily runs, V02, MP tempos, critical velocity and actual threshold
IMO, if i could only have one, it would be sub-threshold medium length intervals
99
harb retweeted
this stuff is so tiresome. jp morgan lead new york city yacht club to two successive cup victories, rockefeller raced trotters through central park, ellison made sailboats fly
you all are falling victim to the pastiche of martyrdom without any of the glory. get over yourselves




















