I didn't see any details, but I'm hoping that you can easily import password lists. I just set up 1Password, I'd hate to have to input each individual password again
I just got 1Password in the last ~6 months, and have noticed this. Specifically when logging into AWS with IAM user. Maybe its just me, but I haven't been able to add more than one username/password to a login which is frustrating, specifically impacts my example above.
I was at a Cavs playoff game and they had these as well. I would guess they will become more popular. I've seen at some games, I think a Browns regular season game, you had to download an app and then give them access to your flashlight which I was not interested in. So this is a nice alternative to create the desired atmosphere without gaining access to users phones
> remember the 1000 Indians that were watching people shop at the Amazon Go stores?
> Amazon told us it was all "sensors" because it fit their company narrative to do so.
I think you, like many others, fell into the trap of just looking at the headline. The 1k people were labeling training data, not watching you shop like a puppet master..
I'm a little skeptical. I grabbed a small dataset and asked ChatGPT to create a linear regression equation and then did the same in excel. Perhaps they estimate values for beta differently, maybe ChatGPT is secretly a Bayesian, but the output I got was different. It was close, which I will say is impressive. Also, maybe I missed it, but I think it would be valuable to add to the prompt to provide the estimates for the parameters. It seemed as if they sort of just took the prediction values, computed MAE, and took the estimate for what it said. As I type this, I do wonder if you could use it to compute estimates for the parameters iteratively, average them, and essentially compute a bootstrapped estimate for the population parameters..
This is pretty exciting. I haven't used Polars much, but GPU support could be something that pulls me over to it. I do hope that Polars doesn't just build with Nvidia in mind, hopefully they'll expand to other architectures over time as well
I haven't listened to Jensen speak before, but am I the only one who thought the presentation wasn't very polished? Not a knock on anything he has accomplished, just an observation that sorta surprised me
I'd argue that the innovation with these models is making smaller. Just throwing compute resources to make a model with more parameters is easy and doesn't really expand our knowledge. IMO, larger and larger LLM's aren't that impressive. Being able to shrink that model down, retain its accuracy (to a degree) and be able to run it on smaller hardware is impressive and will more likely lead to AI/ML being intertwined within people's day-to-day
OpenAI/Sam Altman seem to be trying to do a speed run to becoming unlikeable. Starts off with good intentions of making 'Open' source models available for the benefit of everyone only to change course once they see a market available for the product. There was the debacle of firing him and then re-hiring him a week later, and now he is looking for funding from people known to violate human rights. It is honestly impressive
The first point you made is what myself and my team have struggled with. We have built simple apps for customers to upload a document to interact with and having the state be accurate to the document that the user uploaded was a major pain. Felt easier at that point to build something from scratch than to keep messing around with streamlit's session states
They won't have something at that size because as you pointed out, it is still huge. But depending on how they are used, smaller parameter models may be better for specific on-phone tasks that start to make the size of the model not a problem. GPT4 is so large because it is very general purpose with the goal seeming to be to answer anything. You could have a smaller model focused solely on Siri or something that wouldn't require the parameter size of GPT4
The headline is implying that each individual is being tracked as if for performance reasons, but its a bit different. And is it really that different than the data that companies already keep on individual messaging? The only thing really different from what I can see is that they now have a tool to try and understand the vast amounts of data vs. combing through message logs more manually.
Kids can know to close the door, intend to close the door, but still fail to realize they didn't close the door. They're kids, it happens. And its a simple feature that can help realize the door is still open. It's not that deep..
I'll give a slightly different take on remote work and it has nothing to do with the productivity argument. I go into the office two days a week because I enjoy it. I get to connect with coworkers much better than the days I am at home. I'm able to chat about sports/shows we watch, as well as what is going on in their life outside of work. This is a benefit which has really no tangible benefit to the company, aside from me having more of a connection to my employment. But there is a benefit to this for me. I like the hybrid approach so that I get my days to be social, but then have days to recharge on my own.
On the development side, I also think there are benefits to being in the office. I sit right by my bosses, bosses, boss. He's very busy, but we still have small talk. When it comes time to promotion, there is very real benefit to having that relationship, whether you like it or not. Now I am more goal oriented where I see promotions as a sign of success for myself. Not the case for everyone, and there is nothing wrong with that.
Now the thing I hate about the conversations of working in the hybrid/remote world is the emphasis on making remote workers feel included. I have nothing against people who choose to work fully remote, but a benefit to being in the office is that its easier to connect with people. That's why I do it. Just like there are benefits to working from home, this is one of the benefits of working in office. I feel like this often gets overlooked and people are surprised that they feel disconnected when they aren't in the office. I have become good friends with my coworkers who also go into the office, and barely talk to the who don't.
Are you looking to test different prompts on a set of questions? I have not found anything that specifically does this. When I have tested different prompts, it has just been with a short script iterating through the questions/prompts. Could be a fun project to build something where all you do is add the prompts you'd want to test, but the challenge that first comes to my mind is creating the question set to be used.
As I write this, when you say you want to test prompts, are you looking to test the system prompt on a set of questions or is the prompt the question and just asked in different ways?
How are you guys assessing the accuracy? Sure, LLM's aren't going to be 100% accurate. But if I pick up a Spanish textbook, there is a right answer 100% of the time (okay, aside from publishing errors)
Citadel isn't making the bulk of their alpha from simply finding an edge, its making trades in dark pools based on trade information they receive for executing trades. The individual trader won't be able to make trades at the speed required when finding small price discrepancies, nor would they have the capital to really earn anything.
I find the field of TinyML very interesting. It's one thing to be able to throw money and compute resources at a problem to get better results. But creating solutions that have those constraints I feel will really leave an impact
Do you have sources as to why they are lies? I'm not doubting you, but simply saying they're lies is just as disingenuous as the people making the statements in the first place
So this is just a wrapper thrown around ChatGPT? Inconsistency in punctuation leads me to believe it was done quickly, without a lot of thought and as money grab?
I think its a bit of a stretch to lump exercise simply into medical/health advice. I wouldn't tell you to go on YouTube to diagnose an illness you may have. But OP asked for guides on at home workouts and I think that YouTube is more than sufficient and safe for learning about what different exercises may be available.
You may find something by searching YouTube. I've found that there's a wide variety to the amount of equipment needed and at worse, just pick things out of different videos based on the equipment you have
> “Some of the facilities figured out way too late, ‘oh we should probably be checking how much we’re paying for electricity per month.’ That’s pretty basic when 30-40% of the cost of the crop is electricity,” he says.
> “Those things were just overlooked because investors will pay for it, it doesn’t matter,”
I'm always amazed when startups operate like this. I'm equally in awe when they get capital and VC's don't notice this either. I guess the later could be naive and just assume that the tech required to make it works requires this, but it is still baffling that nobody thinks its a good idea to get one of your bigger costs down