Podcasts about Alakazam

Erik Dobell is doing what many performers dream of doing - running his own theater show. Join me as I dig in with Erik to find out the ups and downs of a theater show.This week's Trick Talk is on Auto Chair by Alakazam. By the way…whenever you're ready, here are 5 ways I can help you grow your magic business to book more shows at higher fees: ·         Grab my FREE business-building tools for professional working magicians. I'm offering SIX free resources to help you improve your online presence. From SEO strategies that almost NO magicians are using, to a social media strategy that can save you time and heartache, to valuable Google Ads strategies, these resources will help you book more shows at higher fees. Get your free copies HERE. ·         Get my website video training for under $10. In this 2-hour video training, I'll reveal exactly what your website needs if you want to succeed as a professional entertainer in the 21st century. Get all the details HERE.  ·         Get a complete business-building plan handed to you on a silver platter. Not sure how to move your magic business forward? I'll analyze your complete magic business (website, market, competition, and more) and give you a complete game plan for getting your performing business to the next level. Find out how HERE. ·         Work with me privately. Need help in multiple areas? I can help you with your website, direct mail, email, or other marketing strategies. I also offer consulting services for crafting magic routines or even entire shows. Want to find out more? Shoot me an email at cris@theprofessionalmagicianclubpro.com. ·         Check out my extensive line of magic routines and marketing products I have several professional routines, including Serendipity, marketing courses, books, and other resources to boost your shows and make booking shows easier and more profitable. Check out my line of products HERE. 

“I would like to see a robit made out of bras.” This time Richard brings a bonus Brooke to talk about witches. He also re-enacts one of our discussion movies. Alakazam.  0:00 -- Intro5:22 -- The Wizard of Oz1.37:50 -- Bedknobs and Broomsticks2:06:17 -- Bewitched2:21:41 -- Contact information2:25:48 -- Awards and rankings3:20:00 -- Future business3:30:37 -- Outro, and outtakes (ooop#28-gas5) Hey! Be sure to watch Fiend Without a Face, The Invisible Man, and Prey for next time! Hey! We have a Patreon (Ours, Ours, & Ours)! Hey! DON'T leave us a voicemail at (801) 896-####! Hey! Shop the Zazzle store! Hey! Hear In Memoriam! Hey! Hear Fantasy Murder Love Triangle! Hey! Hear J.R. Watches Star Trek for the first time! Hey! Hear Richard on Turning Trekkie! Hey! Hear Andy on Kiss Your Franchise Goodbye! Hey! A wizard did it! Hey! Spoon with Angela Lansbury! Hey! Check out the alternate ending of The Wizard of Oz! Hey! Subscribe in iTunes! Hey! Check out the Facebook page and vote on the next category! Hey! Check out Jon's YM&T Letterboxd list! Hey! Check out Roy's YM&T Letterboxd list! Hey! Email us at yoursminetheirspodcast@gmail.com! Send new topics! Send new theme songs!

veronika and kyle saw inside out 2, they test their friendship, and talk some real magic stuff. MERCH!!: https://veronikaiscool.com/ Patreon: https://bit.ly/nevermindpatreon     veronika: @veronika_iscool https://www.instagram.com/veronika_iscool/ kyle: @kylefornow https://www.instagram.com/kylefornow/ nevermind: @nevermindpod https://www.instagram.com/nevermindpod/ we're almost really good at this.

It's return guest season here at Latent Space! We last talked to Kanjun in October and Jonathan in May (and December post Databricks acquisition): Imbue and Databricks are back for a rare treat: a double-header interview talking about DBRX from Databricks and Imbue 70B, a new internal LLM that “outperforms GPT-4o” zero-shot on a range of reasoning and coding-related benchmarks and datasets, while using 7x less data than Llama 3 70B.While Imbue, being an agents company rather than a model provider, are not releasing their models today, they are releasing almost everything else: * Cleaned-up and extended versions of 11 of the most popular NLP reasoning benchmarks* An entirely new code-focused reasoning benchmark* A fine-tuned 70B model, built with Meta Llama 3, to identify ambiguity* A new dataset of 450,000 human judgments about ambiguity* Infrastructure scripts for bringing a cluster from bare metal to robust, high performance training* Our cost-aware hyperparameter optimizer, CARBS, which automatically and systematically fine-tunes all hyperparameters to derive optimum performance for models of any sizeAs well as EXTREMELY detailed posts on the infrastructure needs, hyperparameter search, and clean versions of the sorry state of industry standard benchmarks. This means for the FIRST TIME (perhaps since Meta's OPT-175B in 2022?) you have this level of educational detail into the hardware and ML nitty gritty of training extremely large LLMs, and if you are in fact training LLMs of this scale you now have evals, optimizers, scripts, and human data/benchmarks you can use to move the industry forward together with Imbue.We are busy running the sold-out AI Engineer World's Fair today, and so are unable to do our usual quality writeup, however, please enjoy our show notes and the excellent conversation! Thanks also to Kanjun, Ashley, Tom and the rest of team Imbue for setting up this interview behind the scenes.Video podTimestamps* [00:00:00] Introduction and catch up with guests* [00:01:55] Databricks' text to image model release* [00:03:46] Details about the DBRX model* [00:05:26] Imbue's infrastructure, evaluation, and hyperparameter optimizer releases* [00:09:18] Challenges of training foundation models and getting infrastructure to work* [00:12:03] Details of Imbue's cluster setup* [00:18:53] Process of bringing machines online and common failures* [00:22:52] Health checks and monitoring for the cluster* [00:25:06] Typical timelines and team composition for setting up a cluster* [00:27:24] Monitoring GPU utilization and performance* [00:29:39] Open source tools and libraries used* [00:32:33] Reproducibility and portability of cluster setup* [00:35:57] Infrastructure changes needed for different model architectures* [00:40:49] Imbue's focus on text-only models for coding and reasoning* [00:42:26] CARBS hyperparameter tuner and cost-aware optimization* [00:51:01] Emergence and CARBS* [00:53:18] Evaluation datasets and reproducing them with high quality* [00:58:40] Challenges of evaluating on more realistic tasks* [01:06:01] Abstract reasoning benchmarks like ARC* [01:10:13] Long context evaluation and needle-in-a-haystack tasks* [01:13:50] Function calling and tool use evaluation* [01:19:19] Imbue's future plans for coding and reasoning applications* [01:20:14] Databricks' future plans for useful applications and upcoming blog postsTranscriptSWYX [00:00:00]: Welcome to the Latent Space Podcast, another super special edition. Today, we have sort of like a two-header. John Frankel from Mosaic Databricks, or Databricks Mosaic, and Josh Albrecht from MBU. Welcome.JOSH [00:00:12]: Hey, glad to be here.SWYX [00:00:14]: Thank you for having us. Hey, so both of you are kind of past guests. Jonathan, you were actually one of the most popular episodes from last year talking about MPT7B. Remember the days when we trained large models and there was 7B?JONATHAN [00:00:30]: Yeah, back when reproducing LLAMA1-7B was considered a huge accomplishment for the field. Those are the good old days. I miss that.SWYX [00:00:38]: As the things have accelerated a lot. Actually, let's do a quick catch up and Josh, you can chime on in as well. So Databricks got acquired. I talked to you at New York.JONATHAN [00:00:45]: Mosaic got acquired, although sometimes it feels like Mosaic acquired Databricks because, you know, we're having a lot of fun being here. But, you know, yeah.SWYX [00:00:52]: Yeah. I mean, you are chief scientist now of Databricks.JONATHAN [00:00:55]: Chief AI scientist. Careful with the title. As much as I would love to understand how Spark works, I'm going to have to defer that to much smarter people than me.SWYX [00:01:03]: Got it. And I don't know about like what you would highlight so far as a post-acquisition, but the most recent news is that you guys released DBRX. Is that the thing that most people should be aware of?JONATHAN [00:01:13]: Actually, that's no longer the most recent news. Honestly, the most recent news, we announced this, but it was at our Data and AI Summit last week. So it was announced among like 100,000 other things, is that we finally released our text to image model, which has been a year in the making through a collaboration directly with Shutterstock. There was a lot of work put into finding a dataset that we were comfortable with working on and trying to build a model that honestly, I felt like I could trust and that others might be able to trust to put out in the world. So that model was released last week. It's unfortunately just available via API due to the fact that the data is quite sensitive and quite valuable. It's Shutterstock's entire business in a lot of ways, but I'm still really excited that there's now a model that is trained on a dataset where the provenance of every single image is known, and it's a damn good model. So I'm really proud of the team on that.SWYX [00:01:55]: Yeah, amazing. Josh, do you have any thoughts on image model questions?JOSH [00:01:59]: That is not my area of expertise, but I was excited to see the release of it last week as well, and very happy that you guys did a nice job on the data side of everything there. So that was cool to see.SWYX [00:02:09]: I think what's unusual is like, I think Shutterstock's doing multiple deals in multiple labs. So what is the Shutterstock model? Like, I guess, is this the house model for Shutterstock? Is this Databricks' version of the Shutterstock model? Like, what is this?JONATHAN [00:02:22]: The way that I would think about it is that Shutterstock is doing an amazing business in AI across the board. Their dataset is kind of widely known to be the best stock photos dataset in the world, the most comprehensive, the biggest. When you think about like, what dataset am I going to train a multimodal model on? You call Shutterstock. And I, at least I've heard in the news, like OpenAI, Google, Meta, Apple have all called Shutterstock and made those deals. So a lot of models have had Shutterstock data incorporated into them. But this is the only model I know of so far where it was, you know, exclusively and specifically trained just on the vanilla Shutterstock data. There was nothing else mixed in. We didn't go and scrape the web and find other data or combined datasets or anything like that. And so this is, in some sense, the house blend. But the other piece is that it's just a dataset where the provenance of every image is known in public. Where did the data come from? It is the Shutterstock collection. That's it. You know, nothing less, nothing more. And certainly being at Databricks, if I've learned one thing, I've learned about enterprise customers and what they want out of AI. And one of the things they ask for most is just, what can you tell me about the data the model was trained on? And here, especially for text to image models, where images are just tricky subject matter, there's been a lot of kind of legal conversation about images, especially. It's nice to just have something where I can point to it and say, you know, if you want to know where the images came from, these are what they are and this is how they got there.SWYX [00:03:36]: I will talk a little bit about Databricks because it's relevant to the rest of today's episode. So Databricks, sorry, I keep misspeaking. It's DBRX.JONATHAN [00:03:46]: DBRX, actually, there's been a pronunciation update. It is now D-B-Rex. So we have decided to add a dinosaur mascot because what model doesn't like a mascot? So literally, I wish I could pull it up. There is a little plush dinosaur that we had made. It's like the world's cutest dinosaur, but it is the official mascot of D-B-Rex. And there's a little dinosaur logo that, you know, you'll probably see around a little bit more because DBRX is a mouthful, but D-B-Rex, like, you know, it's just kind of...SWYX [00:04:13]: Rolls off the tongue. I love mascots. Like every company should have a mascot. And I think Hugging Face got it right. You need an emoji mascot because that's the minimal viable image.JONATHAN [00:04:21]: I probably shouldn't talk at all about, you know, Velociraptor, but, you know, that's a, maybe that's something we can talk about later in the summer. I'll just leave it at that.SWYX [00:04:28]: Okay. That's a hint to names. I feel like your names leak a lot of alpha. So just to quickly cover the headline details, DBRX, as Make Sure Experts model, that's fairly big, 132 billion total parameters, so 36 billion active on any input, pre-trained on 12 trillion tokens of text and code, and did really well on evals to the point where you had to dye your hair blue. That's my high level conclusion.JONATHAN [00:04:53]: Never make a bet with your team two weeks out from model launch, even when, you know, human eval is looking quite bad. Because if you set some bar, even if it's arbitrary and you think there's no way in hell they're going to hit it, apparently money doesn't motivate people anymore. Humiliating their boss motivates people. So Josh, you should really take a hint from this. You know, you cannot pay someone enough money to make up for you dyeing your hair blue.JOSH [00:05:15]: I'll keep that in mind for our next model.SWYX [00:05:17]: It works. So speaking of Imbue's next model, perhaps Josh, you want to actually just say hi to the general sort of latent space audience and talk about what we're releasing today. Yeah.JOSH [00:05:26]: I'm Josh, CTO of Imbue, and we're not releasing the model. We're not releasing the weights, but we are releasing a bunch of different things that should make it easier for other people to make their own models. So I think right now, training foundation models from scratch is like a very difficult, time-consuming, expensive, kind of risky endeavor, especially for smaller companies. And the things that we're releasing hopefully make that at least a little bit easier. So the things that we're releasing fall into kind of three different buckets. One is infrastructure and scripts for dealing with the kind of hardware and hardware failures and understanding how well is the actually lowest level of thing actually working so that you can actually do your training at all and at a reasonable speed without having to constantly restart, etc. So infrastructure and training scripts. A second set of things is around the evaluation. So after you've trained it, like how well is this actually working and how do you know how well it's working? We're releasing a whole bunch of different data there, a new benchmark about code, reasoning, understanding, as well as our own private versions of 11 different open source benchmarks. So things like pool queue or ANLI, where we've gone through and kind of cleaned up the data as much as possible by looking at all the ones that models get wrong or that are flagged for ambiguity and also our own kind of private reproductions of those where we've done like a kind of clean room black box, like, okay, this is what the data set is supposed to be. Here are some examples. Let's make our own version of this to make sure that there is no data contamination, etc. To make sure that we're actually, you know, not testing on train. And then I think a final thing that we're releasing there is around 450,000 human judgments about ambiguity and question quality, which we used in the process of cleaning these evaluations and we also hope will be helpful for other people training kind of similar models. And then the third thing is CARBS, our hyperparameter, our cost-aware hyperparameter optimizer, which was especially helpful for being able to experiment at much smaller scales and then scale those experiments up to the much larger scale kind of on the first try without having to retry it. You don't want to be training, you know, 10, 20 different 70B models. You really want to get these larger modelsSWYX [00:07:30]: right on the first try.JOSH [00:07:30]: And so the ability to kind of tune things very precisely and learn scaling laws, not just for, you know, the like data and flops, but also for learning rate and all the other hyperparameters and see like how should you scale these things up was extremely valuable to us as we were training the larger models. Yeah, that's a lot of stuff.SWYX [00:07:49]: Yeah, exactly. So there's a bunch of stuffJOSH [00:07:50]: we'll have to go through all of it.JONATHAN [00:07:52]: Yeah, I just want to throw in how excited I am about this. This is the stuff that nobody ever talks about. That is the difference between success and failure in this stuff. Like, can you get your cluster to run? Can you get software on your cluster? Can you figure out what broke? Because fault tolerance is still not really built into any of the fundamental primitives of training models. And so if something breaks, you have to go figure out what broke, your job stops, you have to restart your job. It is a nightmare just to get to the point where anything can train on the cluster. A basic MPI hello world that has the GPUs talk to each other is hard enough, let alone actually training a model, let alone getting good performance out of the GPUs, let alone actually getting a model that converges to anything interesting. There's so many levels of things you have to accomplish. This is the kind of stuff that matters. I think to a point that Josh made earlier, before we got on here, there are plenty of weights out there. Nobody's released this.JOSH [00:08:46]: Yeah, that was part of the motivation actually is that there are lots of other things that are complimentary, but I have not seen nearly as much discussion about some of these other things that we think are pretty important. I mean, in some sense,SWYX [00:08:56]: I'm very excited to have Jonathan on because this is a little bit, you're a bread and butter with Mosaic. And I think you've released some part with Composer. And I think it's just really interesting to see like a different take, basically a full stack take that's kind of open source today.JONATHAN [00:09:18]: Yeah, it's really kind of, it's been an ordeal to figure this out. And every time something changes, whether it's a new GPU or even a new driver update, you get new creative errors and new things go wrong. And, you know, we've dealt with the weirdest things from, you know, our InfiniBand cables getting stolen from the data center twice, like in boxes before they arrived at the data center. Like, you know, Porch Pirate basically had stolen our InfiniBand cables back when those were hard to come by. To like, you know, weird recalls of switches to like the strangest stuff has happened. I have my favorite GPU failures I've seen, like ones where the GPU doesn't fail, it has a correctable memory issue and the memory correction causes the GPU to become a straggler and hold up the whole job. Like weird stuff happens and figuring out how to not just identify all of that, but then eventually productize it, is in some sense, the entire story of Mosaic and now Databricks in terms of our ML offering. Really, the thing we offer is we have gone through this suffering and figured out how to even productize that. It has been a pain in the butt.SWYX [00:10:20]: Yeah, it's a lot of work.JOSH [00:10:20]: I think my favorite failure was GPU is just giving wrong math. Like if they give errors, great, because you can see the errors, but if they just give you the wrong math back, not so fun.SWYX [00:10:30]: When did they give you wrong math?JOSH [00:10:32]: Like literally you could just, you know, add two things. For example, the numbers come back. They're not the numbers that they're supposed to be.JONATHAN [00:10:40]: I think it's important to say at this stage, just because like it, I think it goes without saying for Josh and I, but it's worth saying here, this isn't to say that like anything is wrong with us. It's not like NVIDIA did a bad job or, you know, Mellanox did a bad job or the like the server builder, the data center operator, the cloud provider, like the million other parties that are involved in building this. We are running these insane chips that are huge and complicated and built on tiny transistors at insane frequencies with insane heat in data centers that for the most part, were not built remotely for this kind of power or heat and have been retrofitted for this. Like failures happen on a good day with normal CPUs. And this is not a good day and not a normal CPU for the most part. It's fun to joke about all the weird things we see. This is not to say anybody's done anything wrong. This is just kind of part and parcel of working on a massive cluster running at multiple megawatts of power at a time.SWYX [00:11:32]: It's crazy. Yeah.JONATHAN [00:11:33]: So optical cables, like all sorts, like everything.SWYX [00:11:37]: I'll take the opportunity to start going to the sort of infra piece. There's just like a description of the infra just to give people a sense of what we talk about when we talk about massive clusters. So I'm just going to read off the blog post here. This post is about one cluster that has 4,092 H100 GPUs spread across 511 computers. They use unified fabric manager nodes, which manage the infinite band network. And you talk a little bit about your networking. Is there anything unusual about this setup that you'll call out to people?JOSH [00:12:03]: Yeah, actually this particular cluster is a little bit non-standard. The normal, like vanilla setup for these large clusters as vanilla as it can be is what's normally like a 127 node cluster. So closer to like 1024 GPUs instead of 4,000. Here we have a larger cluster. As you start to get into the larger clusters, the networking becomes a little bit more custom. It's a little bit more, it's a little bit trickier. It's a little bit more difficult to get these things to all be able to talk to each other at the same speed. And so this has, in this particular case, this is a three tier network architecture instead of two tiers, kind of the normal one. So most of the clusters are a little bit smaller. As you get to even larger scales, then this becomes even much more complicated,SWYX [00:12:43]: much more expensive.JOSH [00:12:43]: So we chose this particular scale, kind of knowing our own workloads and kind of what we wanted to do. This was kind of the right size for us. But yeah, I think it's not exactly vanilla already. It's already getting into kind of the custom territory.SWYX [00:12:54]: So my understanding is that there, and is there any part of this that comes with the Voltage Park deal that you guys had? Is that part of the hardware that you got from the deal with them?JOSH [00:13:04]: Yeah, so we worked really closely with Voltage Park to set up all their clusters and infrastructure and everything and kind of decide even like what to order, how should the networking work? Like we were very involved in kind of the construction and bring up of this. And that's what this post is about, is about that process of like bringing up all these, there's like different clusters in different places of different scales. So in this particular post, we're talking about this one 4096 GPU, but there are other clusters that they have as well. And we were very closely involved with figuring out the exact architecture and kind of the trade-offs that go along with picking, you know, those exact components. You really don't want to like place the wrong order because it takes months to get it and it's very expensive. So yeah, we were happy to help out with that.JONATHAN [00:13:43]: And then your bit of good cables get stolen.SWYX [00:13:44]: Yeah, yeah, exactly.JOSH [00:13:47]: We wanted to make sure that we ended up with compute that would work for us and that would also work for their other customers. And so we kind of helped design something so that we would get exactly what we were looking for. We knew that these kinds of details would be super important and that getting down to the level of the hardware and like having these good scripts and everything was going to be a core part of like actually getting this to work. I'm very glad that we did that. I don't think that most companies kind of take that full stack approach, but for us, it certainly paid off.SWYX [00:14:12]: Yeah, it's basically sort of built to spec. It's interesting that relationship because you usually, for the rest of us who don't operate at your scale, we take whatever we can get from cloud providers, but you are basically co-designing from the single machine up. And you described that a little bit. Do you want to take us through the process that you described here?JOSH [00:14:27]: Yeah, so for the actual, like the blog post and kind of bringing these machines online.SWYX [00:14:32]: Yeah.JOSH [00:14:32]: So yeah, I think the process, as we have it broken down in the blog post, there's kind of a few different layers. First is like getting the individual machines to work at all and then getting the machines to actually be able to talk to each other. So getting the InfiniBand networking to work and then getting to a point where, you know, not just the machines are working and they can talk to each other, but everything is actually working correctly. There's a big gap between like it's working at all to it's working perfectly correctly. And then after you have all this stuff working perfectly correctly, nice and healthy, then now you get into kind of the software data, like training issues. And then after that, you're still not done. Like now, even once you're training at full speed, things are going to fail over time. Things are going to change. There's going to be new, you know, firmware updates. Like how do you kind of deal with this change and flux over time without going crazySWYX [00:15:16]: and pulling your hair out,JOSH [00:15:16]: trying to like reproduce things or understand why there were regressions. And so there's a lot of work to kind of automate the infrastructure tooling as well. And kind of the first step, like bringing these things online in the first place, you know, you have hundreds of machines at this point. So you don't necessarily want to be like walking around with like a CD-ROM or a USB drive, like plugging it in with your keyboard, like hitting next, next, next on the OS install. That's not how this works. You do that for one machine. And then you use, we use this thing called Metal as a Service to bring up all the other machines. So it's a kind of server that can kind of install the operating system on these other machines. So most like when you're talking about these machines, like each machine is, you know, on the order of hundreds of thousands of dollars. So they usually come with a kind of out-of-band management interface as well. So they don't, they have their InfiniBand networking. They have their normal 100 gigabit per second Ethernet networking. These are like dual, redundant, et cetera. And then you also have this extra out-of-band management network. So you can log in and you can see like the boot screen or you can see the blue screen of death. You can like get in there and actually see what was wrong, which is pretty fun. And it makes it like possible to automate a lot of this work. So the beginning of that, and the blog post goes into much more detail about like exactly how we set these up and kind of the other errors that we ran into. When you're bringing these online, you'll definitely have failures. Even if they all worked in the factory, they get shipped, some parts come loose, something fails, something goes wrong. So when you're bringing them online, there'll be some that don't quite work for all sorts of reasons. As you start to be working with machines at this scale, like if something happens one in a thousand times, you're like pretty likely to see it. And so you can get pretty rare, weird things, especially since we had fairly early builds and fairly early versions of this hardware. Like these are some of the like first machines that were ever produced, some of the first GPUs. So you've got some extra special things there. We definitely worked with Dell, for example, on making fixes in the firmware level to be like, okay, like this thing is wrong. Like we need to update this at the firmware to like actually fix this particular thing. So we worked pretty closely with Dell and Nvidia. Yeah, that's what I'm saying. Like this stuff gets complicated. And the thing is like, you know, taking a step back, the whole reason we're doing this, right, is that we knew that this was going to be complicated. There would be these kinds of failures. And if we're just using, you know, AWS or some other cloud provider, these errors are still gonna be there and you're gonna have no way to know and no way to debug this and no way to diagnose what's going wrong. And so we would much rather be able to like call up Dell and say, hey, this isn't working. And they're like, yep, okay, cool. Let's debug it together. Oh, I see. Yeah, cool. We'll ship a firmware update and actually fix this for you. That was a much better experience than like, great, just magically fails. I guess we restart and hope that that machine goes away. Like that's not a very good place to be. So yeah, that's kind of the first place is getting to a place where like GPU training is working on your single node machines. You can observe stuff. We have tons of tooling around like, you know, Prometheus and all sorts of other tools for understanding what's going on in these machines because you don't want to be like logging into each one and looking at the temperature or something you really need to have tooling to collect all these metrics, et cetera. Unfortunately, all of the scripts that we have for this are like for this entire cluster and for all this infrastructure are a little bit like special purpose for our particular thing. So it's not that every script that we have, it's not that you can just like take this and plug this in. Even if we did open source all the tooling that we have, you'd still have to do like a lot of work to open source it. What we are releasing is as many of the things that we can that are going to be useful for other people. You're still going to have to have some way of kind of managing these things, making your own like logging aggregators, et cetera, et cetera. So that's kind of bringing them up to the like, you know, the single nodes that are working. From there, it goes into, I'm happy to keep going if you want. Well, I just want to leave the opportunity for JohnSWYX [00:18:53]: to comment if there's anything that's different from how he runs things.JONATHAN [00:18:57]: Oh, I mean, all I'll say is I'll endorse this and say this s**t is hard. Like this is really, really hard. And, you know, I have a special props to, you know, the folks in Vue because they were building this from the ground up. You know, at Databricks and at Mosaic, we typically work with cloud providers because some of this stuff is just, there's too much to handle. It's complicated. There's a lot to deal with. And this doesn't even get into things like physical security, you know, securing power if you're the data center operator. Like this gets infinitely complicated and you have to abstract somewhere. Like, you know, and then you get to the folks who are literally building their own custom chips and like, good God.SWYX [00:19:36]: Like, oh my God, that's, you know,JONATHAN [00:19:38]: if you're one of those folks, you're having, you know, pour one out for the infra people at some of the AI chip startups who are having a really, really interesting time right now. But this stuff is really hard. And I don't think we talk about it much because there's so many other things that are hard. But the other hard things, I think everybody's becoming pretty familiar with at this point. This is something that I don't think there's ever really been a comprehensive discussion of, at least not that I've seen.SWYX [00:20:00]: Yeah, so my impression is that you guys, Mosaic, have your own software for sort of spinning up and down machines, just like Imbue had to build. But Imbue probably, it sounds like Imbue, you guys went fuller stack. I don't know how to describe it. Like Mosaic is not working with Dell on like their firmware.JONATHAN [00:20:21]: No, no, we're typically working with like, you know, pick your cloud provider on their Dell firmware or what have you. Like, it's kind of, I think one of the things, I don't know, Josh, you can correct me on this. It's kind of impossible if you're doing training to not go all the way through the entire stack, regardless of what happens. Like somehow I'm still chatting with cloud providers about power contracts, even though the whole point of dealing with the cloud provider is not to have to think about power contracts. Somehow I'm still asking them about which InfiniBand provider they used this time to see if this is part of the bad batch of cables I encountered on that cloud provider or what have you. Or like, we're still talking about a firmware update from pick your provider. You can't not do this. It's convenient that they have data center staff who are worrying about what to send back to which provider when, and they have people who can go and wait for the InfiniBand cables so they don't get stolen outside. But, you know, it's kind of, it's impossible not to really go full stack if you're thinking about the infrastructure at all. I don't know, Josh, correct me. No, I think that's right.JOSH [00:21:17]: That's what we expected from the beginning as well, is that we would inevitably have to get into the details here. And I'm glad that we kind of just planned for it. I think it made it a lot easier from our perspective to have direct control over this. Instead of having to go to the cloud provider that goes to the data center, that goes to the supplier, we could just go direct to NVIDIA or DellSWYX [00:21:37]: or the data center,JOSH [00:21:37]: whoever was responsible and be like, hey, this thing needs to change. And they're like, oh, okay. Yeah, that is our responsibility. Great, we can fix that. So it was just a lot easier for us to fix these bugs than if we had to go through an extra layer of email.SWYX [00:21:48]: Something we discussed in the pre-show was that you had a rule of thumb for your cluster of reliability. You say here in the post, by and large, you expect around 3% of your machines to break every week. So you're basically going to turn through all your machines in a year.JOSH [00:22:04]: As it says in the post. So that would be true if it was a uniform failure like that. But as it says in the post, it's usually these kind of problematic nodes. And to be clear, that is the number that we've heard from other people is like they're having about 3%. I don't think we're experiencing failure rates that are that high. I think ours is actually quite a bit lower than that, probably because we've taken the time to like dig into a large, maybe larger number than we should have of these failures and get to the root cause of it and be like, oh, okay, like that's exactly what's going wrong.SWYX [00:22:33]: How do we fix this?JOSH [00:22:33]: How do we prevent this from happening? How do we make automated checks for this so that if it does happen, it just goes back to whoever owns that particular part of the process and they can fix it immediately.SWYX [00:22:43]: And that's part of what you're also open sourcing, which is the health checks, right? You got the NIC health checks, GPU health check, this space health check, Docker D message. I don't know what that is.JOSH [00:22:52]: That one is just a lot of stuff.SWYX [00:22:54]: Yeah.JOSH [00:22:55]: That one is one where we realized that actually like when these machines boot, sometimes they wouldn't actually boot cleanly all the way. Or when they rebooted, they had problems that they didn't have when they were working before, which was kind of frustrating. Like usually if you restart your computer,SWYX [00:23:08]: it gets better.JOSH [00:23:08]: Here you restart. It did not get better.SWYX [00:23:10]: It got worse.JOSH [00:23:10]: That was very frustrating. So this health check looks at every particular line we've ever seen from the boot, like in D message, like every single log line that your computer emitsSWYX [00:23:21]: and says like,JOSH [00:23:21]: have we ever seen this before?SWYX [00:23:23]: Is this expected?JOSH [00:23:23]: Is this in the right order? Or is there something out of place? If there's anything out of place, let me say, okay, great. Like now it goes into this, like longer, more triage list of like, all right, great. Like, is this acceptable?SWYX [00:23:33]: Should we flag this?JOSH [00:23:33]: Like, should someone take a look at this? So we're looking down at a very, very granular detail level, what's happening on these computers to make sure that nothing is out of place. And that's critical because without that, if you're running your training, as Jonathan said, and this thing is slow, like what are you supposed to do? Right?SWYX [00:23:49]: Like you really,JOSH [00:23:49]: you really want to be very certain that like all 4,000 of these GPUs are working like they're supposed to.SWYX [00:23:54]: We know that.JOSH [00:23:54]: And so if it's slow, it's because like we messed up the config or something else and not because of this earlier thing that's like really hard to detect in software later.JONATHAN [00:24:01]: Yeah. I think the, I'm just curious to ask,SWYX [00:24:03]: like, you know,JONATHAN [00:24:03]: suppose you were to set up another, let's say another H100 cluster and it were at a different data center. And instead of the vendor being Dell, it was super micro or what have you. How much of this would be repeatable? And how much of this would you have to redo? I, you know, I genuinely don't know.SWYX [00:24:18]: A decent amount.JOSH [00:24:19]: I think it would go a lot faster the second time. I think there's lots of learnings that we had. And also the blog post,SWYX [00:24:24]: you know, yes,JOSH [00:24:24]: we are releasing the health checks, releasing some scripts, but a lot of the valuable stuff is also in the blog post itself, in the details and kind of the, you know, the learnings that we've had and the sort of errors that we run into. We tried to as much as possible surface those to other peopleSWYX [00:24:36]: could learn from thoseJOSH [00:24:36]: and avoid the same mistakes or failures as well. But I think it would go a lot faster.SWYX [00:24:41]: Although, yes,JOSH [00:24:41]: there would certainly be some things that'd be a little bit different. I mean, there'd probably be different CPUsSWYX [00:24:46]: or whatever,JOSH [00:24:46]: but I think a lot of that stuff is less,SWYX [00:24:49]: it's less,JOSH [00:24:49]: that's the like, that's less variable. I think most of it would apply the second time around. Although I'm sure next timeSWYX [00:24:56]: we're building one,JOSH [00:24:56]: it'll probably be, you know, at a scale that's 10x as big with a different chip or something like this.SWYX [00:25:00]: And then who knows?JOSH [00:25:01]: Yeah, with Kinect X8,JONATHAN [00:25:02]: that will have its own fun behavior and all that good stuff. Yeah.SWYX [00:25:06]: Perhaps there's something that people don't discuss about, and you don't even talk about this in the blog, but I always wonder is what is the timeline that's like kind of reasonable for this amount of work, at least the initial stages? And also what does the team composition look like for setting up a cluster, right? Like what are the mix of skills that you typically would require to get all this going?JOSH [00:25:27]: I'm, I can't really speak to typical. One thing I am very proud of is how much we accomplished with such a ridiculously small team. Like our infrastructure team is like, you know, fluctuates from week to week, depending on like how many things are on fire and how much we need to build. But it's like between like three and six people, like it's small. It's not like some huge team of like tons and tons of engineers. But those people are very, very good at what they do. And so that has allowed us to get a lot of mileage out of out of these things. I think it's not that we're building everything, right? It's not that three to six people build this whole thing. I definitely want to like, you know, say thanks very much to Dell and H5 and NVIDIA and the other people that have done a lot of the work, like to bring up this cluster, you know, with 4000 GPUs and three tier networking, networking architecture, you have 12,000 cables. So that's 24,000 things that need to be plugged in. Like that's just a lot of stuff to plug in, right? And you don't want to mess it up. Like each one needs to be done correctly. Like it's a little bit loose. Like it doesn't really work.SWYX [00:26:23]: If you break it,JOSH [00:26:23]: you need to replace it. Like there's a lot of workSWYX [00:26:26]: that goes into this.JOSH [00:26:27]: Yeah.SWYX [00:26:28]: And then, you know,JOSH [00:26:28]: that's just like that's it. That's if you were to do everything right the first time.SWYX [00:26:32]: And if you didn'tJOSH [00:26:32]: have to fix anything. But inevitably, you know, you will have to replace something, which means like taking all the wires out, pulling the thing out, taking all the GPUs out, going and fixing some cable, putting it all back correctly, putting it back in, doing this every time. So there were a lot of people at Dell, NVIDIA and at H5 that all helped a ton with this stuff. I don't know the exact size of the Dell team. It also fluctuated over time.SWYX [00:26:55]: Yeah, excellent. And then, you know, you so you have all the hardware set up and now you're firing it up for a single node. There's a long description that you guys have about just like monitoring the MFU, right? And what each situation might look might be indicative of. One of the most interesting things to me that I saw from here is like, you know, if training immediately starts off at 60 to 80% MFU, something's wrong.SWYX [00:27:24]: But like, you know, like what what are like, you know, some anecdotes or, you know, notable scenarios here that you might you might call out as maybe counterintuitive or super interesting.JOSH [00:27:36]: There's just so many of them. I mean, one of them, which I think is probably pretty common, like common knowledge by this point. But like we did have a sort of likeSWYX [00:27:46]: which one was this exactly?JOSH [00:27:47]: I think for the MFU, like gradually getting worse over time. I think that one, when we saw that the first time we were like, what the heck is going on? Like, why does it get just like a little bit worse? This is so strange. Like, what is it getting lazy or tired or something? Like, is it heat? Like what's going on? And in this particular case, it was memory fragmentation. Because you have hundreds of machines, they're doing garbage collection slightly different times. And then they get slightly further apart and slightly more and more jittered until eventually they're all happening kind of at random times. And just like really messing up each one of your steps. So you just turn off garbage collection and call it a day, basically,SWYX [00:28:20]: to be honest.JOSH [00:28:20]: There's other things you can do if you want to be a little bit more sophisticated about it. But you can also just manuallyJONATHAN [00:28:25]: have it all garbage collect on some interval. Like that's what we've done. We just have a garbage collection callback that just runs. But I've seen the exact same thing.JOSH [00:28:33]: Yeah, yeah, exactly. So I thought that one was kind of funny. And we did trace that one down and look and we did find the actual call. Like, again, this goes to like having good tools. So we had really good tools where we could look at a bunch of like actual traces in C and be like, OK, cool. This is the thing that's taking a lot of time. Or like, you know, this is the thing that doesn't quite line up here. Like, oh, I guess it's garbage collection. OK, cool.SWYX [00:28:52]: Interesting.JOSH [00:28:52]: Yeah, let's just try taking it off.SWYX [00:28:54]: OK, great.JOSH [00:28:54]: That's what it was. Now we can fix it. So for each of them, like basically bugs are not hard if you have good tools. But if you don't have good tools, bugs can be very, very hard. So similarly for like heat, another thing that we saw was like, oh, you know, the CPU is getting throttled. OK, well, it's easy to see if you're monitoring the CPU throttling or monitoring the heat. If you're not monitoring that, it's really hard to know why it's just suddenly one of them is going slower. I noticed also in the pieceSWYX [00:29:17]: that you mentioned FSDP with 0.3. Actually, we met, I went to iClear and Guanhua from the DSP team was there presenting 0++. I was wondering if you want to make any call outs to, you know, particular open source or open library or open whatever implementation teams that were super helpful in your process. I think we ended up actuallyJOSH [00:29:39]: pulling from a whole bunch of different ones to pull things in into our own particular pipeline. So we use things from NVIDIA's, you know, Megatron stuff. We use stuff from probably DeepSpeed. I think we pulled in a bunch of different pieces from a bunch of different places. So it was really nice to see all these working open source like examples. I think I really appreciate all the effort that has gone into actually tuning these things because you can tune them, but it's a lot of work to like tune this stuff and do all this stuff from scratch. It's really nice to have like a working example. I think those are probably the two biggest ones, DeepSpeed and Megatron alone, but there are probably other ones as well.SWYX [00:30:13]: Is there a particular thing in the ecosystem where you would call out as like, you know, there should be something here that is open source, but like it's not really, it's like everyone kind of builds it on their own. I want to say something with the file system because everyone talks about the file system eventually.JOSH [00:30:28]: The file system actually was,SWYX [00:30:30]: I mean, we did somethingJOSH [00:30:31]: kind of dumb there. Like we have our own sort of local mirror so that we can, you know, like a crappy version of S3SWYX [00:30:38]: that's local,JOSH [00:30:38]: but it's just a pretty simple script, right?SWYX [00:30:41]: Like I think we run likeJOSH [00:30:41]: a little web server that just like serves files and then, you know, it can upload themSWYX [00:30:45]: and download them.JOSH [00:30:45]: Okay, great. And part of the reason we did that is that our internet connectionSWYX [00:30:50]: in the beginningJOSH [00:30:50]: was not the like full speedSWYX [00:30:52]: one that we wouldJOSH [00:30:52]: eventually have. And so we are a little bit more kind of bottlenecked in terms of internet bandwidth. And so we had this. I think we looked at a bunch of services out there like Minio and some other ones, but a lot of these like come with a lot of extra overhead and maintenance. And since we already have so much infrastructureSWYX [00:31:09]: to deal with,JOSH [00:31:09]: we kind of didn't want to, you know, bring in a whole other like cloud provider, virtualize something, something.SWYX [00:31:14]: We just wanted something simple.JOSH [00:31:14]: So we went with that, which has been quite helpful. Like our toolsSWYX [00:31:19]: are usually quite simple.JOSH [00:31:19]: It's like Bash and Python and SSH and Docker. Like we'd like to keep things simple so that's easier to debug, like less layers of infrastructure, less layers of abstraction, make it a lot easier to work with. Like we don't use Kubernetes,SWYX [00:31:30]: for example,JOSH [00:31:30]: and we just directly launch these things. And it's just been much easier to debug this way. One tool actually that does come into mind that I will call out is Kraken from Uber. That was great. We love that tool. We were a little bit skeptical. What is it?SWYX [00:31:44]: I'm sorry. Yeah.JOSH [00:31:45]: So Kraken is this, yeah, it's a distributed like Docker registry, basically, that uses BitTorrent to like transfer things between the machines in a sort of nice optimal way. Like in the very beginning, the naive way is like you have this one Docker registry, which was outside of the cluster. So every time we change an image, you know, there's many gigabytes that each of the 500 machines needs to download.SWYX [00:32:07]: So that just takesJOSH [00:32:07]: a really long time. So what this thing does is like just one of them downloads it and then like they all sort of broadcast all the pieces to each other. And it was just like a really nice, fast way of getting these images down. And it was very robust.SWYX [00:32:19]: Like there's a lotJOSH [00:32:19]: going on under the hood, but I think it's a pretty cool tool that we haven't really had any bugs with it at all. Amazing.SWYX [00:32:26]: Yeah. I mean, that's all my questions, I guess, for the info piece. I don't know if, John, you had something that you were sort of burning to ask or.JONATHAN [00:32:33]: No, all I can say is just sameSWYX [00:32:36]: in a lot of places, like, you know, and they're done thatJONATHAN [00:32:38]: seeing this plus one. I think the one big difference, you know, perhaps in philosophies is we've tried to basically standardize on as much commodity stuff as possible, just because, you know, I think the reason I asked about trying to do thisSWYX [00:32:50]: on multiple differentJONATHAN [00:32:50]: pieces of infrastructure is like, I think we're running on like six or seven different clouds right now. And everybody has done something slightly different. And my gosh, the little differences add up as you know, you've seen. And so, you know,SWYX [00:33:04]: our philosophy has been like, whatever the hellJONATHAN [00:33:05]: we can standardize, please let's standardize it. Like vanilla off the shelf FSDB.SWYX [00:33:10]: And like, you know,JONATHAN [00:33:10]: we wrote our own data loader, but we've tried to make that as much of a standard as we can across our infrastructure and in Databricks, because things just start getting really complicatedSWYX [00:33:18]: or like we useJONATHAN [00:33:18]: Kubernetes extensively because it at least gives us a uniform set of APIs. Like that's our hardware abstraction layer to a certain extent for everything else. So it's just, you know, a difference in philosophy there. But otherwise, like, yeah, this stuff is really, really hard. And I feel like we take for granted how much of this, you know, is done for us when you go and you just query chat GPT, for example. Like, oh my God, everything going on underneath that, you know, it's kind of a miracle that the machines boot up, let alone that you can like query a giant language model that's probably doing inference across multiple machines and was trained across thousands of machines. Like, you know, minor miracle.SWYX [00:33:54]: Yeah, it is an awesome amount of power that we invoke with a single API call that we take for granted these days. It's absurd. Yeah, I mean, like Kubernetes, like that point about Kubernetes, I will say as a former AWS employee, like it seems like it would be ideal for imbue to at some point make it more abstracted or agnostic because you're going to want to, you know, replicate your setup. We do have our ownJOSH [00:34:19]: sort of replacement. It's just a much simpler version of Kubernetes. Kubernetes is really designed for running services, not for running experiments. Like that's not its like main architecture. And so for us, like we have everything that's like, cool, you're going to run an experiment. So you want it to run to completion, right?SWYX [00:34:34]: OK, great.JOSH [00:34:34]: Like the primitives are sort of built around a slightly different style. And that makes it a lot easier, like just a lot simpler to fit that the nature of like these machines are going to disappear. They will need to be rebooted for infrastructure upgrades. They will like something will happen to the GPUs. Failure is like baked into this as like a core part of our infrastructure. So it's not that we don't have an abstraction. It's that it's a sort of simpler, more tailored abstraction for the particular work that we're doing.JONATHAN [00:34:58]: Yeah, I think it all depends on what your goals are. And like, I think the challenge in a lot of the deep learning stuff right now is that people are trying to like, people often build things that are more complicated than necessary to get the job done. And the complication is the enemy of everything. You know, don't use a fancier parallelism strategy than you have to. Don't use a fancier set of libraries than you have to.SWYX [00:35:18]: Don't do anythingJONATHAN [00:35:18]: that you don't have to do because it's hard enough as it is. Like, don't overcomplicateSWYX [00:35:23]: your own life.JONATHAN [00:35:23]: Don't try to bring in more tools or more fancy architecture tweaks if you absolutely don't have to.SWYX [00:35:29]: Like getting to the minimumJONATHAN [00:35:30]: necessary to get the job done. And it's really tempting to want to try to use everything. So like, I totally understand that one.SWYX [00:35:37]: I think the last piece I'll maybe call out is that I'm just going to weave this in just because I see the opportunity to do it. Are there any infrastructure shifts that need to be, that need to rise because of changing architecture? So I think, for example,SWYX [00:35:57]: you're announcing a dense model, a 70B dense model, whereas John just worked on DBRX and the image-to-text model, which presumably has different bottlenecks.JONATHAN [00:36:10]: That's correct for us. You know, we train both dense and mixture of expert models. The one we happened to, you know, kind of get permission to open source was a mixture of expert model. And those models are very demanding when it comes to network bandwidth, at least if you're training them in kind of FSTP 03 style, where there's just a lot of parameters getting shuffled back and forth. And your ratio of kind of compute to amount of data that you have to shuffle back and forth becomes a lot worse because you're now, you know, you're only using a fraction of the parameters for every token instead of all the parameters. And so we had to really push the envelope on getting all the stuff to the right places on time. And so actually the networking part of DBRX was the single hardest thing, I think, of the entire process. Just get MOE training, working at scale across a big cluster. We still managed to, I think, do it all with commodity parts, which was very exciting. You know, we were using FSTP and we eventually used HSTP so that we could have HSTP as a version of FSTP where you have multiple smaller replicas and you're doing data parallel within those replicas. And that helped a lot with network latency issues that we were running into just because we were transmitting so much data, you know, for every single part of the process. I think it actually, like, it was instructive for how Google designs their hardware and software together personally. Their training, as far as I understand, using kind of a 03 style of training and have been for a while. They also train mixture of expert models. TPUs have a very different network bandwidth to compute ratio. They have a lot more bandwidth just objectively. And TPUs per chip tend to be a little bit less compute intensive and have a little bit less memory. You know, it's just a different design choice. So the ratio of flops to bandwidth is very different. And that means that it's much easier for Google to be able to pull offSWYX [00:37:54]: some of this stuff.JONATHAN [00:37:54]: They also have interesting, you know, Torus style network architecture or Torus style, like, literal network architectureSWYX [00:38:00]: is not like the model,JONATHAN [00:38:00]: but the network.SWYX [00:38:02]: Is this the sort of block attention? I forgot what you call it. So this is just more or the,JONATHAN [00:38:07]: yeah, this is more, not the ring attention, but these are the ring all reduces. Like you have three different dimensions of rings because they kind of put you in these three dimensional Toruses from what I understand. And so like, you know, Google's infrastructure in some sense is kind of, I wouldn't say built for this, but maybe the way that Google trains models is built for a slightly different bit of infrastructure they have. And it's kind of neat to think about that. You know, as one thing that I think NVIDIA announced for, you know, for, for both the GH200 and the GB200 is this hybrid networking where you'll have blocks of NVLink network chips. I think for the GB200, I think it's like groups of 72 GPUs will all have NVLink to each other. So higher bandwidth, then you'll have normal networking of some kind, InfiniBand or Rocky or what have you between these blocks. And that's kind of a, you know, it's a change due to the fact that, you know, it's hard to build really high bandwidth networks over very large groups, but it is now a blocked networking. And you have to think about how you architect your model and your parallelism differently. You also have to think about fault tolerance differently because it now matters where you lose a GPU, whereas it didn't before. So, you know, it's, it's, it's just all really interesting and really fun speaking personally, but it's going to mean new nightmares when we all move to that generation and have to think about, you know, new versions of these problems.JOSH [00:39:20]: As you go up to larger scales, it gets quite different. Like right now, you know, if you're experiencing, let's say, for example, you experience a GPU failure every day, that's fine.SWYX [00:39:31]: Just restart.JOSH [00:39:31]: If you make your thing 24 times as big, now it's once an hour. Now it stops being quite as easy to just restart, right? So now you have to kind of break, like bake in this sort of redundancy that you didn't have before. So I think as you go up in scale, you end up running into like a lot of really interesting problems that also inform the, the actual like design. Yeah, I mean, as an orchestration guy,SWYX [00:39:52]: this is why I always emphasize like very cheap storage or very fast storage. So you can checkpoint more, but I don't think that's probably not the best solution to for fast, you know, training.JONATHAN [00:40:05]: Which works fine when you're doing language and then you move to vision or video. And then, you know, you have multi petabyte datasetsSWYX [00:40:12]: and getting, you know,JONATHAN [00:40:13]: cheap, fast multi petabyte storage starts to bite. Like I've certainly encountered issues where the literal data center where my GPUs were did not have enough, you know, object store to fit the datasets that people wanted to bring into that data center from whichever users were, were trying to bring them in. And then you get to a wholeSWYX [00:40:31]: different world of hurtJONATHAN [00:40:31]: where you have to keep your data in a different region because the region is just out of storage. So things get fun really fast.SWYX [00:40:39]: Speaking of vision, Josh, actually, you know, Embu is an agents company, but you're only, you're announcing a text-only model. What, where does, where does the vision side come in?JOSH [00:40:49]: I think we've actually done a lot of work in the past and people can see kind of our blog posts about sort of self-supervised learning and some other kind of vision-related stuff in the past as well. So we're very familiar with, with that stuff. But I think our main focus right now is on kind of, as we say, coding and reasoning. And there, there's certainly a visual component to some problems. But, you know, it's not necessarily required for all problems. And actually we found that for most of the kind of like code writing and, and reasoning problems that we care about, the visual part isn't really a huge important part of it. Sometimes if you really need to, you can maybe describeSWYX [00:41:24]: the thing.JOSH [00:41:24]: There are other like, you know, multimodal models that you can use off the shelf to sort of plug in for those particular piecesSWYX [00:41:30]: that you need, right?JOSH [00:41:30]: Like if something is driving a browser or whatever, like you can sometimes get away with not having to have that baked into the original model. So our folk were, you know, in a sense, we kind of do a lot across the stack. We're working on our own infrastructure and pre-training and RL and fine tuning and products and everything. But in another sense, we're very narrowly focused on the application side. So all of the stuff across the stack is kind of going toward a very particular purpose. And so that particular purpose right now doesn't really need vision. So we think that people are going to make all sorts of really cool image modelsSWYX [00:42:00]: like Jonathan, right?JOSH [00:42:00]: And all sorts of interesting multimodal models into the future. We'll let them go do that. That's great. We'll take advantage of that, partner with those people in the future. And right now we're really focused on kind of the core reasoning and coding capabilities and aspects of the model.SWYX [00:42:14]: I wanted to go into carbs since that's kind of the next layer of the stack. We talked about carbs in the first episode with Kanjin because you've actually had a blog post about it like a couple of years ago. Maybe let's introduce it.JONATHAN [00:42:26]: Has that been a couple of years now?JOSH [00:42:28]: No, it must have been at least one year. Hopefully it's not multiple years.SWYX [00:42:32]: Sorry, I'm counting AI time. Yeah, yeah. Yeah, I was going to sayJONATHAN [00:42:35]: you're making me feel really old right now.SWYX [00:42:39]: I count everything before the generally intelligent rename as like, you know, prehistory. Yeah. And now sort of modernity, right? So I actually thought carbs was more about hyperparameter optimization in a sense of like sort of parameters, hyperparameter search. Whereas, you know, when you introduced it, especially in this blog post, it's more about scaling laws and predictability of like, are we sort of in the right ballpark before we scale things up? Maybe sort of recount the history of carbs.JOSH [00:43:10]: Yeah, so it really is a little bit of both. So carbs is, it's maybe a backronym, but it's for cost aware Pareto region Bayesian search. So this is about technically how it works, but carbs is like, you know, we like pastries and stuff.SWYX [00:43:26]: So great, why not? But the point is thatJOSH [00:43:29]: it's a cost aware hyperparameter tuner. So most hyperparameter tuners, you kind of say, OK, here's this objective function. I want you to make this number as big as possible or as small as possible, whichever direction you want to go. So yeah, just go make this number, you know, as small as possible. OK, so it'll try a bunch of differentSWYX [00:43:46]: hyperparameters,JOSH [00:43:46]: a bunch of different configurationsSWYX [00:43:48]: to figure out, like,JOSH [00:43:48]: how do I tweak your network and architecture, et cetera, to get the kind of best performance I possibly can. That's usually saying, like, you know, almost all of these hyperparameter configurations are, let's say they're all going to use the same number of GPUs or the same number of nodes.SWYX [00:44:01]: So it's going to runJOSH [00:44:01]: for the same amount of time.SWYX [00:44:03]: So you can do that.JOSH [00:44:03]: You can get a number out and that's great. But what carbs does is it says,SWYX [00:44:07]: OK, actually,JOSH [00:44:07]: what if we relax that constraint? What if we say each of these different points, we're going to model how expensive it will be to sample this configuration. So if what if we train with just one one hundredth of the data? Like, how well can we do?SWYX [00:44:19]: What if we trainJOSH [00:44:19]: with one tenth of the data? What if we train with all the data? That way you can understand, like, as we get more and more data, as we spend more and more compute,SWYX [00:44:26]: as we make a biggerJOSH [00:44:26]: and bigger network, how does performance change with these things that change? Like how expensive it is to even explore this data point. So by doing that, we can see the scaling laws for not just, you know,SWYX [00:44:36]: the scaling lawsJOSH [00:44:36]: from like the, you know, Chantilla paper, the scaling laws for all parameters. We can see how does how does the number of layers change with this? How does the, you know, the learning rate change? How do the like, you know, various types of regularization change? So you can see these nice scaling laws. And as you're going across costs, like how should this be changing as you're scaling up your model? So that, coupled with the kind of metric that we chose, which is a very precise way of measuring performance, allowed us to really like hone in on parameters that worked really wellSWYX [00:45:05]: and understand, like,JOSH [00:45:05]: how do we want to scale those up, especially as we're changingSWYX [00:45:08]: things about the network?JOSH [00:45:08]: Like one of the things that we did is we used a custom tokenizer. As we change this tokenizer, changes a bunch of other things about the model. So how should we scale up this entirely new tokenizer? Like no one has ever made a model this large with this tokenizer before. And so how do we want toSWYX [00:45:22]: change all these things?JOSH [00:45:22]: Harps kind of shows you, like, look, as you change these parameters, like these other ones are kind of dependent on this.SWYX [00:45:28]: Like this is the, these areJOSH [00:45:28]: the relationships between them. So you can better understand, like, OK, if I'm going to scale this up 10x or 100x, like, where do I want to be? I can only go so far. And so, you know, we did run, like, I think maybe it was like a 14b one or somethingSWYX [00:45:40]: like that to check.JOSH [00:45:41]: But and so we had a bunch of like 1b or 14b and then at 70b. I don't think we had a, I think we just did like one at 14b. So you can, we get to check that like, oh, is this on the curve? Like, is this where we expect? It was like right there. So then great, go on to the next one. Yeah, I mean, that makes a lot of sense.SWYX [00:45:56]: I wonder if, so one of the key questions, and correct me if I'm wrong, but like usually people do search or do their evals just based on loss. But you actually evaluate based on, you know, the sort of end state evals that people might expect, like HellaSwag and Lombata, whatever. What is the norm here? Is there a norm?JOSH [00:46:20]: Yeah, I don't know if there's a hundred percent.SWYX [00:46:21]: I don't know. I only see loss on most people's reports.JOSH [00:46:25]: I think it's easy to, like, loss is very nice because it's very precise. It will tell you, like, very fine grained differences between like really small changes in your hyperparameters or network architecture. Whereas, especially at the smaller scales, if you're looking at like accuracy, it's very noisy. Like it might be zero or a hundred or like, you know, fluctuating by like 10 or 20 percentage points, which makes it really hard to tell, like, did that change actually mean anything? So our loss is sort of a combination of these two. Instead of saying, like, let's just look at perplexity, we say, let's look at perplexity on the tasks that we care about for multiple choice questions effectively.SWYX [00:47:00]: So we're saying like, yes,JOSH [00:47:00]: this is formulated as a multiple choice question, and we're going to look at the, like, you know, the loss of perplexity for this particular answer token. And that ends up being something that's like both targeted to what you actually care about and also very precise. The nice thing about this though is that it's independent of the data that you train on. One thing that's annoying about perplexity or about loss is that as you change your data set, this is really obnoxious because now it fundamentally changes your loss, right? And so you can't tell, like, how do I tweak my data set? But because we have this held out evaluation dat

This week the lads were joined by podcast royalty, Andrew Gode from the We Got The Chocolates podcast. The boys go through the best and the best of magic round, for another healthy dose of rugby league fun and games. --- Send in a voice message: https://podcasters.spotify.com/pod/show/josiah-alchin/message Support this podcast: https://podcasters.spotify.com/pod/show/josiah-alchin/support

ALAKAZAM! Time for some magic, this week we're talking about Now You See It! Tune in as BenG and Kelsey discuss the weird escape room puzzles in this movie, the logistics of kids producing a reality TV show, and if using real magic during a magic competition show is cheating.

This Saturday, Craig is joined by the legendary Peter Nardi, Harry Nardi & Jamie Daws to talk the latest development of Alakazam Magic, Alakazam Unlimited! Alakazam Unlimited is a monthly service that will allow you to gain unfettered access to over 150 extraordinary magic effects!

It's the run-up to Blackpool Magic Convention, and there are a LOT of dealers going. The question is what are they all bringing, and how best can you plan your trip? Today, Craig talks to Peter & Harry Nardi about what he is bringing to Blackpool! Thanks for watching.

This week, we continue our deep dive review of the Harry Potter series with "Harry Potter and the Prisoner of Azkaban". These kids are growing up fast. Wassup with Marge? Is Dumbledore ok? Why is magic communication so nonsensical? Ya boys are Siriusly Black and siriusly back. Tune in for this and more in Issue 92 of The Mighty Metahumans! For advertising opportunities & inquiries, please email: TheMightyMetas@gmail.com For the VIDEO edition of this podcast, sign up for our Patreon! Patreon: http://www.patreon.com/TheMightyMetas Twitter: http://www.twitter.com/TheMightyMetas Please take a min to leave a review / rating if you enjoyed the episode --- Send in a voice message: https://podcasters.spotify.com/pod/show/the-mighty-metahumans/message

The local chapter of a historic, national magicians organization — once led by Houdini himself — is facing a shortage of members. Now its ranks are rallying with events to woo new magicians.

Alakazam! Alakazoo! I get to interview Scorpius Malfoy from Harry Potter and the Cursed Child woo hoo! Erik Christopher Peterson is currently starring on Broadway in the Tony Award winning play Harry Potter and the Cursed Child as Scorpius Malfoy (the son of Harry Potter's fiercest rival, Draco Malfoy).In this interview Erik is Baring It All About:Magic SpellsBeing on BroadwayStage MishapsHow he guards his energy/timeSo much moreHarry Potter and the Cursed Child plays at the Lyric Theatre in NYC. More on Erik Christopher Peterson:Erik Christopher Peterson is a NYC based actor/creator with a mission to use storytelling as a catalyst for increasing empathy within his audience, and himself. Currently he can be seen as Scorpius Malfoy in the Broadway Production of Harry Potter and the Cursed Child.​Erik holds a BFA in Acting from Webster Conservatory where he studied contemporary acting for the stage and screen, as well as a wide array of classical styles including his personal favorite: Shakespeare. ​Erik is also a huge baseball fan: Go Rockies! Special Thanks:Rachel Jensen, FINN PartnersTheme Song by Bobby CroninPodcast Logo by Liam O'DonnellEdited by Adam RothenbergConnect with Me:Website: www.callmeadam.comFacebook: @CallMeAdamNYCInstagram: @CallMeAdamNYC

Wir reden heute über die 10 meistverkauften Kunststücke vom Großhändler unseres Vertrauens. Da sind einige Überraschungen dabei. Versprochen. Genauso wie gute Laune und Fachsimpelei vom Feinsten. Alle Infos in den Show Notes, über 15 000 Artikel beim Secret Magic Store.

On tomorrow's new bowl of Question Crunch, Marky Make Up talks about the amazing products and content that they create. Answering questions about being a cosmetic mechanic, reporting at Drag Con, and what new costume Mario should wear. #usedmediaorg #podcast #questioncrunchpodcast #marykaymakeup Listen on Spotify Apple Podcast Question Crunch Site QuestionCrunch: Instagram: @QuestionCrunchPodcast Facebook: https://www.facebook.com/questioncrunch Host: Jimmy Purcell @beenbettercomic Intro Jingle: Nathan Gonzales Producer & Editor: Angie Dee @angiedeerivas @USEDmedia --- Send in a voice message: https://podcasters.spotify.com/pod/show/questioncrunch/message Support this podcast: https://podcasters.spotify.com/pod/show/questioncrunch/support

There's a serious problem in the community garden! The beloved apple tree is in a sad state, and Redge worries there won't be apples for snacking, pie baking and cider making in the fall. Abracadabra! Alakazam! In swoops Zinnia Fig riding on a cluster of wild onions to declare the tree is experiencing the dreaded DROOP. Can the Folktale Forest's favorite hedge witch use her practical earthly magic to cast out its doldrums and bring on the blooms?Episode webpage: https://jonincharacter.com/zinnia-fig-and-the-droop/     SUMMARY: Jonathan and Redge are enjoying a beautiful day in the community garden. They notice the apple tree, which should be blooming already, doesn't look happy so they call upon Zinnia Fig, the Folktale Forest's hedge witch, who says the tree is experiencing a sad state known as “The Droop.” Zinnia explains that sometimes trees need to feel their way through sadness and offers support and encouragement to the tree. They fertilize the ground and speak kind words to the tree, which starts to show signs of improvement. The flowers on the tree begin to bloom again, and Zinnia assures them that with care, the tree will fully recover. Grateful for Zinnia's help, Jonathan and Redge express their appreciation, and Zinnia performs a final incantation to bring joy to their day.PARENTS, TEACHERS, HOMESCHOOLERS:  Overall, the story promotes themes of friendship, patience, positivity, acceptance, collaboration, and a deep appreciation for the beauty of nature. It encourages kids to cultivate a sense of wonder and respect for the environment while fostering a supportive and caring community.  If you liked this story, you may also enjoy Episode 50: The Magical Doorway, where Jonathan enters the land of Once Upon a Time and meets his best hedgehog friend, Mr. Reginald T. Hedgehog: https://jonincharacter.com/the-magical-doorway/Dorktales Storytime Podcast website: https://jonincharacter.com/dorktales-storytime-podcast/CREDITS: This episode has been a Jonincharacter production. Today's story was written and produced by Molly Murphy. All characters are performed by Jonathan Cormur. Sound recording and production by Jermaine Hamilton at Hamilton Studio Recordings.We're excited to share, Food for the Future: Sustainable Farms Around the World, by Mia Wenjen and illustrated by Robert Sae-Heng. It's a children's book about 12 unusual farms where people grow our food in ways that care for our planet. We know your young readers will love it! Published by Barefoot Books: https://miawenjen.com/food-for-the-future-sustainable-farms-around-the-world/Support the showREACH OUT! Tweet us @dorktalesstory Email us at dorktalesstorytime@gmail DM us on IG @dorktalesstorytime Newsletter/Free Resources: https://bit.ly/dorktalesplus-signup One time donation: https://www.buymeacoffee.com/dorktales Original Music Available on Bandcamp: https://dorktalesstorytime.bandcamp.com/music Now, go be the hero of your own story and we'll see you next once-upon-a-time!

Today, Craig gets the opportunity to talk to Chris James, one of the biggest upcoming names in magic. Chris has been creating a range of abstract effects that are releasing with Alakazam and Murphy's Magic, and saw great success with the free release of Pop Wild! He is well know for taking already established tricks and giving them a new fresh twist that makes them way better! Thanks for watching.

Harry Nardi is an incredible magician, magic creator, Britains Got Talent Finalist, Social Media Influencer, Alakazam Magic Marketer and Product developer and a true rising star of magic. Harry returns to the Scrapbook Magic Podcast so if you haven't heard his first interview way back in 2020, then go and listen to that first!In this episode, Harry shares his incredible journey of growth, aspiration and discovery, sharing with us everything about his life over the past 3 years, through the pandemic, heading to Vegas and Blackpool on his own as an Alakazam brand ambassador, the decisions made when developing the Alakazam website and tons more. This is a real peek into the future of one of magics biggest companies and in the first podcast with Harry, I mentioned how it was a joy to witness his growth through magic but now that joy is amplified ten fold. It's been a privilege to see his journey and now you get to experience it with me. @harrynardimagicianwww.alakazam.co.ukSupport the showFind out more about the creator of Scrapbook Magic here: www.jamiedaws.co.ukIf you want to help support the Podcast, please consider becoming a member of VIP Study. VIP Study is a purpose build learning platform for performers and magicians with over 40 hours of bespoke content from building props and scripting to bizarre magic and tons of tricks! Visit www.vipstudy.co.uk to find out more!

We have a whole host of new Pokemon TCG cards to go through, from the next set (named Paldea Evolved) and beyond!   Lots of great discussions to be had, don't miss out!    

This episode has a couple of interesting things, but Adam the Alakazam really does take over the episode... The AhdurinCast Feed - https://theahdurincast.podbean.com/ The map: SPOILER_map.PNG (1604×809) (discordapp.com) Patreon: https://www.patreon.com/notascratch Website: www.notascratchcast.com AhdurinRegion Twitch: https://www.twitch.tv/theahdurinregion Podcast Twitter Link: https://twitter.com/NotAScratchCast DM's Twitter Link: https://twitter.com/AhdurinRegion Discord link: https://discord.gg/kfhBdkTDnX AhdurinRegion YouTube: https://www.youtube.com/channel/UClRPG86yjATufmxvrgcfAVA Muhammad/LumiShen YouTube: https://www.youtube.com/channel/UC4QJYoKfnhEIr61eYflfkWg Other Show: Ahdurin Region Sports: https://anchor.fm/ahdurin-region-sports Other Show: Ahdurin Talks Sports: https://anchor.fm/anirudh45 The AhdurinCast Feed - https://theahdurincast.podbean.com/ Music by GlitchxCity: https://www.youtube.com/user/GlitchxCity/ Music by TableTop Audio: https://tabletopaudio.com/ Music by Bit by Bit Sound: https://bit-by-bit-sound.itch.io/ Promo in episode: N/A Make sure to follow, subscribe, and do all those important things that you should do when supporting a podcast! DM: Anirudh Kyle: Muhammad

Harry Mandel has been performing magic most of his life. Inspired by the Amazing Randi on Wonderama and Mark Wilson on his Magical Land of Alakazam back in the early days of kids television shows. Harry has always been intrigued by Carnival hucksters and pitchman, as well as the escape artistry of Harry Houdini. For more about his magical life, visit www.mandelmagic.com

This episode, we welcome E.Bursonson (Ed Brown) to the podcast and start with a discussion of comics we grabbed at Alakazam comics in Irvine, CA (4:00) before we discuss editing multiple comic scripts and story planning (24:40), doing the hard work on pages, so final editing can happen (37:00), and working on commissions and compensating for you own time (48:30).

Your all access, behind the scenes insight into the current happenings of the Magic industry. Where Craig & Lloyd will thrust themselves deep inside your ears every Wednesday to rant, dissect and give you their raw & unfiltered views on the art of Magic So lock your door, take out your deck and brace yourself for The Magic Podcast. New episode every Wednesday at 9PM GMT. www.themagicpocast.co Disclaimer: the following podcast contains strong language, Welsh jokes, magician's testicles & more. But let's be honest, you all love that stuff!

In this episode, I discuss the GoFest Finale and how all the GoFest experienced ranked overall, the introduction into the Season of Light, all 4 forms of Deoxys returning to raids, the upcoming Psychic Spectacular Event, go over whether Mega Alakazam is worth raiding, and finally, finally, touch on the Clefairy Commotion event.Socials Opening Song: Music by Alex MakeMusic from PixabaySeptember Content GraphicHow Good is Mega Pokemon Series:AbomasnowAerodactylAmpharosBlastoiseCharizardGengarPidgeotScizorSlowbroVenasaurHave you checked out my guest appearance on Purified Podcast? Click below to listen!Episode with Purified Podcast

In a crossover event of the millennia, the Bellhop Boys have the hosts of The Alakazam with Alex and Sam Podcast on the Bells 'N Whistles couch! They discuss podcasting methodology, Hollywood magic, and Elmo. Let's hope everyone leaves with the correct co-host! Follow Samantha on Instagram: https://www.instagram.com/salmonthabrooks/ Follow Alex on Instagram: https://www.instagram.com/realsteinberg/ Find Alakazam with Alex and Sam on Instagram: https://www.instagram.com/alakazampodcast/ And check out their website: https://www.alakazampodcast.com SUPPORT US BY BUYING MERCH: https://my-store-7405526.creator-spring.com/listing/bellsnwhistles Please leave 5 stars and a review on Apple Podcasts and follow us on Spotify! https://rss.com/podcasts/bnwpod/ Subscribe to Bellhop Productions on YouTube: https://www.youtube.com/channel/UCC1dr278XrRkoZReO4QoBiA @BellhopProductions on IG and TikTok Check out our brand new channel 3 Ways to Survive: https://www.youtube.com/channel/UCfBAtsmIKSfute9zVzRjfwA For video game content, subscribe to Another F*cking Let's Play: https://www.youtube.com/watch?v=SdwDCr8px9Q SECRET FUNDRAISER TO BUY US MIC'S: https://gofund.me/da965bac Love you!

For første gang på Retropol får et av Pokémon spillene sin helt egen episode. Det passer dermed bra med et spill som er bra for multiplayer: Pokémon Stadium på Nintendo 64 hvor vi snakker om blant annet Kid's Club, Free Battle og diverse Pokémon som Gengar og Alakazam.

We are excited to have on good friend and supporter and advocate of our industry, Tony Prestipino from Artifex Dental Laboratory (https://artifexlab.com/). Tony comes on to talk about how he learned about the profession, having a brother who is a prosthodontist and another bother in the lab business, getting into implants, and taking an All-on-X procedure and patenting the procedure. It's hard to find anyone so willing to help anyone willing to ask. Information and how to register for the Race For the Future 8.0 August 28th (https://dentallabfoundation.org/news-events/race-for-the-future/) Information and registration for the Chicago Triathlon (https://www.chicagotriathlon.com/) Register for the FREE Education Day on August 27th (https://dentallabfoundation.org/education-day/) (COME TO THE DINNER!) Donate to Barb or Team Voices From the Bench!!!! (https://fdlt.memberclicks.net/donor-form#!/) Whip Mix (https://www.whipmix.com/)'s new VeriWhirl (https://www.whipmix.com/products/veriwhirl-dental-3d-printed-parts-cleaner/?product=technical_resources) resin cleaning station is the ideal piece of equipment to use in the 3D printed resin post process. It's oscillating, multi-speed stirrer produces a tornado-like vortex every 30-seconds, and guarantees efficient, effective and powerful cleaning whether the units are individual or still attached to the build plate. The two alcohol baths make an effective step-wash system. It cleans more efficiently and there is less alcohol needed, since it is re-used for both a fresh bath and a dirty bath. The VeriWhirl's intelligent design offers features such as Mode, Time, and Start/Stop button display which gives the operator full and automatic control of the cleaning process. A mesh basket used in the wash container makes it easy to keep track of small printed parts when cleaning. The affordable unit's one-year warranty ensures peace-of-mind for the owner. Visit whipmix.com (https://www.whipmix.com/) or call (800) 626-5651 for more information about this great new product. ExoCAD is excited to present their next Insights event October 3-4, 2022, on the sun-drenched island of Mallorca, Spain. Learn how DentalCAD (https://exocad.com/our-products/exocad-dentalcad?utm_campaign=Insights%202022&utm_source=Introtext%20Insights%20sub%20page%20EN&utm_medium=DentalCAD%20product%20web%20page%20EN&utm_term=link), ChairsideCAD (https://exocad.com/our-products/exocad-chairsidecad?utm_campaign=Insights%202022&utm_source=Introtext%20Insights%20sub%20page%20EN%20ChairsideCAD&utm_medium=ChairsideCAD%20product%20web%20page%20EN&utm_term=Link) and exoplan (https://exocad.com/our-products/exoplan?utm_campaign=Insights%202022&utm_source=Introtext%20Insights%20sub%20page%20EN%20exoplan&utm_medium=exoplan%20product%20web%20page%20EN&utm_term=link) can streamline and improve workflows. Connect with friends and colleagues in the industry while collecting CE points* during a full program of software education, partner shows and international guest speakers. Enjoy the exclusive welcome party, featuring fantastic music and food. Join us in making Insights 2022 an event to remember! Register at exocad.com/insights2022 Did you know Asiga (https://www.asiga.com/) has over 500 validated materials on their open material system. And it's growing everyday? By harnessing Asiga's proprietary layer monitoring technology with its smart positioning system and integrated internal radiometer, as a laboratory, you will be able to produce any indication you desire. Whether models, splints, temporaries, or even permanent crowns. Your investment will be future proofed by Asiga's rugged engineering. Providing you with a fast, accurate, and repeatable machine, with a reputation that is time tested in the laboratory industry. If you would like to learn about Asiga's machine or material offerings, please visit the website at asiga.com or contact your favorite dental reseller. Special Guest: Tony Prestipino, CDT.

Fiz a minha magia mas ela não funcionou! Por que? Será que essa magia vai funcionar? Vou tentar para ver se funciona! Vamos elucidar o que faz uma magia ter sucesso ou ser fadada ao fracasso nesse episódio! Vem com a gente! Alakazam, Simsalabim, Abracadabra! [caption id="attachment_13498" align="aligncenter" width="1080"] Inscreva-se em nosso curso de Firmezas de Umbanda[/caption] Conheça nossos cursos em www.perdidoead.com Este episódio foi transmitido no dia 09 de Junho de 2022, às 21 horas. Siga nossa página no Facebook. Ouça o episódio gravado aqui: Download Link Alternativo Assine nosso Feed RSS Apoie o Papo na Encruza: Sendo um apoiador você nos ajuda a manter a estrutura do blog e do podcast em alto nível. Também irá concorrer a diversos prêmios conforme a sua categoria de apoio. Quer ganhar descontos em nossos cursos, concorrer a sorteios de livros e de Macumboxs? Clica abaixo e nos apoie! =) Seja um apoiador do Papo na Encruza no Catarse e concorra a diversos prêmios Livros Recomendados: Conhecendo a Umbanda: Dentro do Terreiro, o livro de Umbanda escrito pelo Douglas Rainho e lançado pela Editora Nova Senda. E-Book: O Espiritismo, a Magia e as Sete Linhas de Umbanda - Leal de Souza E-Book: Guia do Praticante da Umbanda - Douglas Rainho  Confira TODA nossa Bibliografia Recomendada. Citado no Episódio: Cursos no PerdidoEAD: Um local para conhecer e praticar. Instagram Cova de Tiriri. Oráculo de Exu com Douglas Rainho. Redes Sociais: Grupo do Facebook: Papo na Encruza - PODCAST Facebook do Papo na Encruza Instagram Papo na Encruza (@paponaencruza) Instagram Douglas Rainho (@douglasrainho7) Instagram Luiz Guenca (@guenca) Instagram Macumbox (@macumbox_) Instagram Cova de Tiriri (@covadetiriri) Fale com a gente Caso queira entrar em contato conosco, para enviar dúvidas, comentários e sugestões, nosso e-mail é contato@perdido.co. Muito obrigado todos os nossos apoiadores! Sem vocês esse programa não poderia acontecer!

Roman Armstrong is an incredible magician. He has been performing magic since the age of eight years. You have probably never heard of him, however, he is making HUGE WAVES in the magic industry. Very soon he is releasing his first creation through Alakazam Magic and it is an absolute KILLER!

On this week's episode, Micah and Eric talk about what Fantastic Beasts 4 needs to do in order to keep fans interested in the franchise. Then open our Muggle Mailbag to take feedback on Secrets of Dumbledore. Plus... Alakazam! Can wizards and witches really use spoons to do magic? Main Mini-Discussion: What does Fantastic Beasts 4 need to do in order to be a satisfying story for fans? With where Secrets of Dumbledore left us, what more story is there to tell?  Can the franchise continue to hang its hooks on Potter nostalgia, but then fail to deliver? Do we feel there is enough meat for five Fantastic Beasts films? Could a flashback movie work? It's time to bust open the Muggle Mailbag! What did Grindelwald know about Dumbledore's future when he said: "Who's going to love you now?" Will Jude Law's Dumbledore ever reenact the Half-Blood Prince potion scene? Is Secrets of Dumbledore too gay? Did Dumbledore plan for Jacob to be an "assassin" all along? Why did Leta choose Theseus? Is Newt to blame? Which characters are the best examples from each Hogwarts House? Can you use spoons to do magic? Quizzitch: In Half-Blood Prince, during the cave scene, what does Dumbledore toast before drinking the potion? On next week's episode: we're off for the Memorial Day Holiday, but back the following week with our Secrets of Dumbledore movie commentary! This week's episode is brought to you by Mint Mobile (to get your new wireless plan for just $15 bucks/month and get the plan shipped to your door for FREE, visit MintMobile.com/MuggleCast) and BetterHelp (get 10% off their first month at BetterHelp.com/MuggleCast)!

(listen with good bassy headphones) Marcus Veltri intro (first, second, last clip) https://www.youtube.com/watch?v=t_JrcAOfAp4 00:00 cold open 01:49 Yeti, Pickle, Guitar 29:30 Guitar, Star Wars, Black Hole "You make beats?" https://www.youtube.com/watch?v=BNC68gQluSE05:01 rhyming platypus, cleanser, eternal, onomatopoeia Miami Guerrila Beats: https://www.youtube.com/watch?v=_6yoJTup3nY10:58 miami beach Behind the Beats: https://www.youtube.com/watch?v=JqTMbYvc57g13:41 practice Stevie Knight interview https://www.youtube.com/watch?v=pwn1UsNNmZE17:37 top 5 influences God tier doubletime  https://www.youtube.com/watch?v=Js2Q8tJLgig26:30 Harry potter, Octopus, Jacket First Clipright on time h mac with the freestylerhyme i be coming through to get thatshinewhen i start i'ma blow your mind yeahand i'm going over headskinda like a bald spot yeah as soon as icome in with the lyricshey imma make your jaw drop let's moveimma make his own improve every nightlike it to the grove yeahevery time i show what i'm about when ibe rapping they cover their mouthyeah i'm tearing rappers to pieces icome in hot and they couldn't believe itheyi make them back down you got rat goodsin your background i can see those ikick free flows i come off top when iripyeah every mac every time around manpeople around the planet they can't evenget a grip hold up they love the vibethat i provide all at once everybodylook to the sideyeah imma make his figures go boomwhat's going on someone coming in theroomyeah h man i'm speaking with a purevoice please said the girl in thesweater that is turquoise harry metcoming off the tip of the mantle whenthey come to bars man they know i'messential mad effect i'll be the onethey acknowledge peace to the girl withthe black finger nail polish imma solveit any time i was happy every man youknow they'd be loving what i'm rappingyeahyou know what i'm all about look to theside how does he call it outyeah imma do it like none other yeahi've been calling out the nail polishcolor hey harry mack i be leaving tracksmother ayeand you know i put them under pressureboys that they couldn't even measure allup in the background you got the dresserdrawers i give y'all more i checked herout territoryi explore when i spit it yo it's hard torecordevery time that i beat rapping they cansee we really acting upi'm spitting with the passion i got allmy enemies backing up let's goSecond Clip - Yeti, Pickle, Guitaryeah yeah yeah hey rappers ain't readycoming through shop like machete y'allknow my lyrics is deadly my name isharry compare me to the yetiyeah whenever i'm rapping y'all knowthey're gonna go ham hey scaring y'alllike the abominable snowmani'm coming through and y'all know mylyrics quite pure scary kinda like thehairy eddy with the white furhey yo i'm doing me i hope youunderstand it they popped up in the tentwe were like damn are y'all campingit was like yeah we chilling in thebackyard never thought you'd be thepiano player in rap starbut here we are coming through do itswift yeah find another rapper who coulddo it like thisoff the handle forever we coming throughso easily and back when i first startedtold you ain't no one believing me butnow i make it happen y'all my style isnot for rant kid kinda like yoursituation thisis intense hmac break it down y'allbetter feel mei never hold it down cause i'mdelivering the real meyeah i come with the sick flow a lot ofthem don't even get those but when theybecoming a pringle major flavor i'm likea picklei'm here to make it happen i'm thedopest with the bars yeahflavor like a pickle when you open inthe jar yeah i keep it real and mythoughts are concealed word to picklescause i'm about to give y'all the dealyou feel me i break it down real swiftjust like this get your highlike a spliff i'm really bout to driftmoving through it when i'm on it i'll bedelivering bars hey yomarcus on the chords but he ain'tplaying guitar yeahcheck how we do this man we plucking thestrings yeah for me it really isn'tnothing to sayhey yo i come right off the top when ibe flexing i get strongerwe popped up and she was like hold upare y'all songersi'mma break it down one time h may becoming through with that astoundingrhymeyeah hey yo y'all know i get loose wheni create rhymes peace to both of y'alland the homie up on facetimeThird Clip - Platypus, Cleanser, Eternalyeah coming off the top and i'm spittingthat heat he's like you gotta rhyme overone of my beasts i'm like no doubt manwe bout to get loose i was like oh youmake beats tell me you produce he waslike yeah you know i got the fire i needthree words just to get inspired hmaci'm about to keep it true he was likenah wait a minute i'mma rap for you uhwe about to flip it to my birth so muchpassion my heart really hurts hmag y'allknow that i'll really be a boss as soonas the beat drops i'll be fully goingofffrom the heart imma gladly push i'm ananimal platypus hmeg when i rhyme and ireally get ill like a platypus mouthi'll be getting to the billsget into the money uh i live with everyday and be sunny when it comes to lyricsman i've never been a dummy uh hmacreally about to kill this [ __ ] i'm aboutto make you feel the [ __ ] i'm real withthis whenever i be on it man i'm aboutto leave a rhyme on censored uh y'allknow whenever i be all up on the scenekeep it clean like i came through withthe cleanser yeah hmac man i get it doneuh how to show where i spit it from yeahand my lyrics be the mean type spray abit of cleanser then hit you with ascreen wipe what i mean to say is theytalking i live it hit the screen rightcause i'm always coming through vividhmac with the hd cam whenever i becoming off the top i'm bout to slam yeahy'all know i'm bout to spit it clean andmy whole demeanor is something like soapback in the game just to fill him upwith hope preaching on the microphonesomething like the pope yeah y'all knowi'd be the illest freer first word hethrew was onomatopoeia i was like i'dlove to do it for you boy but theproblem is oh i did that one before hman you know i'm really about to slamhard on mine up here cause my flow boombam how you spend i've been comingthrough and making noise hma rocking forthe girls and the boys and the wholeentire team and the whole entire squadwhen it comes to kicking livers peopleknow that i'm a god yeah they're likehold on wait a minute cause they see meshout out to these dudes mentioning thebeanies hmac man i'm really about tosmash this dude in the back he'd berocking with the glasses hey comingthrough with the freshness ain't theliving rapper on the planet who can testthis rhyming up suggestions y'all knowi'ma burn you every time i grab themlike you know i'm about to murk you backin the day y'all we used to smoke purplenowadays every rhyme i kick is eternalyeah i always be in the [ __ ] when irhyme all my verses are infinite and ido it off the top without no facade iused his word he said oh my god all thisdude is way too real way too ill everytime that i spill ready to build i'm alloffended when it comes to lyrics i'm tooinfinite i've never been timid i step upto the podium i'm going and i'm showingthem exactly what i'm all about it'severy man stay up on my rhyming routeit's all about the time and now i'mcoming with the rhythm i'm making youlisten i'm showing my vision i got youall knotted i do this often keep onrocking till i'm up in the coffin yo getnobody other really do what i do thehomie with the beanie and the braidsyelling outhey man i make it understood over topthe beanie you even got a hood hmacalthough i be the rhyme say a beanie anda hood man my rhyme got layers h manit's off the top check all the way thatthe kid could ride oh when i spit i'd bemaking a laugh now from the front tothat tail up in the background yeah andi'm living my dream right behind thisdude i can see the screen hmac when irhyme i make it look easy so you'reabout to see me all up on your tv yeahi'ma make it known h man man i can't getbusy when i first pop up they like yowizzy now i make their heads been roundtill they dizzy gotta turn and pointinto the tv like did he say that is hereally all typing all his raps here iknow it's unbelievable i said the levelthat i'm on is inconceivable uh y'allknow i come through properly yeah i knowi'm slowing down now a piece to thehomie in the chat who just blessed mewith a name for his soundcloud imma haveto go and keep your beats now h gettingbusy with the freestyle every time he'srhyming they be knowing he's wild i'mabout to hold this [ __ ] down till i'mseen now uh i surround you soon ain'tanother [ __ ] who could clown thedude and you'll know i'll be shiningbright kind of like the colorful lifethat's flashing all around your room ohhmac i got hella freeze i'll be shiningbright like the leds yeah i've beengetting busy in the circle from red tothe orange to the blue green purpleevery different color it all combineswhen i spit i'll be teaching all the artaround i said all these rappers on mylevel i'll be thinking hardly turn onthe colorful lights let's start theparty whatFourth clip - Miami beachhey what's up guys how's it goingmy name is harry mack i'm doing videosfor my youtube channelcan i do a freestyle rap for you guysreal quick yeah no no so can you give melike a random word to inspire myfreestyle umbeach there you go all right we startwith that you can yell them out while igo toookayall rightyeahyeahyeahyeah okay okay okay uh keep a low proand i stay out of reach i'm theprofessor with lessons to teach theylining up just to hear on my speech hadto bring my vibe out here to the beach imake it happen y'all know that i'm sickwith this setting it up and you know shepredicted it mac i come through y'allknow the scope shout out to my girl withthe gold on the hopesfreestyle so my rapping is infinity yeahand y'all know i control the vision damnyour errands looking like the solarsystemyeah yo i can see the sun in the planeth man y'all know that our drive rap iscrazy you got the solar system but younever been spacey holding your attentioncoming over time with the lyricalinvention yeah y'all know that i'm aboutto beat youin miami beach florida yeah i'm about tomake it known cause when i rap i'll beentering the crazy zone y'all know isplit it off top with the flare unlikeyour shades i am never a square back i'mnot breaking down the one cold when i berhyming i'm breaking the rules uh wegonna pull up in the menzies baby in thefancy ass lenses y'all know we do it heygrade she don't even need the sunglassesshe ain't throwing shadey'all know what we're all about anythingthat i see i'm about to call it outmaking it happen i'm crazy we're rappingi never been slacking i'm taking theaction lyrics off the tip and this isabout to make you think she'd be rockingorange sample [ __ ] in the pic every timei rhyme though i do it for the heartshe's been wrecking all right and shereally brings the art yeahall these other rappers hardly ranfitted from the heart yo i knew it fromthe cardiac every time i'm spitting manyou know it's too astounding girls outin miami beast start bouncing againwith the realness hey i'm gonna take thegame on a field tripy'all know we out here working withnever gonnaall around i got love for y'all otherrappers we're about to rise above themall i'ma leave i'm the manuscript forhelp like spanish this is the momentwhere my freestyle's finished let's goFifth clip - Behind the Barswe're gonna try some different youknow rhyme schemes different patternsdifferent placements of the rhyme so youknow a b a brhyme inside a rhyme you knowa a b c c b d e e f now i'm just playingbut uh you know some stuff from poetryclass man we're gonna work on itmikey what were you saying we shouldcall thispractice oh traffic school this istraffic school right herea b a b which is kind of like you'reweaving two separate lines uh twoseparate rhymes in and out of each otherso if you think a shirt then you thinkof wearingthen you set them both up like keep thespeakers blaring because mac is doingwork ask me what i'm wearing i'm in theblack shirt hey yo to blow up in thegame they expect me crazy crazy here'sthe a b a b hold up hey yo i rock y'allknow i'm off the head right theserappers stop just like they at the redlight hey yo smash every time i bringthe vocals these rappers getting gassedthey remind me of a mobile yeah it'sclear the people know i rapping forthhey yo i'm here to show you how we goback and forth uh got what you need i'mgetting plenty live yo i'm adjusting myspeed it's only 25 hey yo i bust theselyrics imma let go heavy like a bus getaround like the metro hey yo stalkingrappers manifesting phlegm hey yo theywalking they something like pedestriansstepping to the mac illinois bust off igot you chilling like you waiting at thebus stop hey yo i drop with no bungeei'm giving not no [ __ ] hey yo i'mfeeling mad hungry let's hit the tacotruck uh cow know that i'm illin thatfellow be never chilling i'll be pushingstraight through them to yellow a dragy'all know i'm off the brain nowy'all heard me make a mistake i [ __ ]up right lots of times when i'm doing mypractice people are like ohwhy is he making mistakes or yo that'sthe first time i really heard him make amistake it's a good observation sohere's what's going on when i'mpracticing it's different than when i'mperforming right so when i'm practicingi'm working on something very specificand i'm sticking to it repeating itagain and again and again and trying tochip away at my weaknesses so in thatcase you're gonna make more mistakesyou're working on the things you haven'tmastered yet yeahy'all know that mac amazes anytime hetouches the mic he blazes kids allacross the planet they bout to cravethis we bout to practice our four barphrases so you need three rhymes likeblurt and berserk i'm out my mind agemac goes berserk turn on my mic cause mybrain's about to blurt yeah y'all knowi'll be rocking the black shirt anytimei grip up on the mic i'm doing mad worknow this is something most of y'allcan't [ __ ] with but it's something thatthe industry is stuck with slam down onthe ground leave the floor rupturednormally it's just two lines and that'sa couplet but we bout to double thatextend it to four hard to the core immabreak them off with more hmac got thebeautiful lyrics the fans adore and wheni'm finished rocking they beg for theencore yeah am i the illest off the tiphold me dead right hey yo we can make ithappen right here on site hey yo everytime i shine with the head bright y'allknow that we chillin right here at thered light uh i told you i'm about to hitthe target anytime the mic is in my palmi'ma spark it lots of rappers rappersbut see mackin is the artist i'm numerouno like the name of the market yeahcompared to other rappers i be sickeri'm contagious with the flow and mybrain works quicker yeah my light stayson strong and never flickers maybe weshould take a pit stop here and cop someliquor all right listen i'm the numberone up on the word scene uh hey yo imight be splitting your spleen yeah hmactell me what the words mean now i'm ineffect moving forward cause it turnedgreen i'm never slipping microphone iswhat i'm ripping well imaginary mic isin my palm but y'all listen to the vibesand divisions while i'm staying on mymission i'm michael jordan these rappersjust assisted scotty kippentraffic school babySixth clip - top 5 rappersYeah, ay, ayAll the way freestyle off top, you know how we doLook, look, look, look, lookCold when I spit it like December windSaid I'm cold with the lyrics like I'm EminemAin't a rapper in the game who could ever just gel meI've been influenced by him since The Slim Shady LPWait, even before that, you know we been the shitPre-Slim Shady LP, his flow was infiniteH-Mack, on it, yo, this is what we always aboutSee, Stevie knows whaddup, he's callin' it outListen, I'ma break it down like thisY'all know I bring the hard flowsWhen it come to lyricism right off topMan, I'm really tryna do it like MarshallMack all up in it, you rappers don't matterYou are not a factor, you are just actorsI'ma do it like Eminem, he be spellin' the initials, Marshall MathersWe about to throw down right hereAnd you know the game is lit for Harry MackSo let him tell me what my name isYeah, we do it like this, throw your hands upYo, would the real Harry Mack stand up?Let's get it, man, I'm throwin' it downI be a king up in this bitch, I need the goldenest crownNo one can really rock this wayYou know that I rock all dayYou know me and the beats be the perfect combo likeEminem to Dr. DreSo let me put it down for all o' y'allYou know that I always scald tracksThey be sayin', "Who's your top five, Mack?"Well listen, you know I get up on the track and then I kill staticNumber one on my list, he made IllmaticYeah, you know there's no rapper goin' past meLike old school Nas 'cause my vocals hella nastyNext level when I do my thing, that's for sureI'm stillmatic like Nas in the orange velourYeah, check the way I break it down, I'm makin' people listenFeel like Nas up in the Queens projects with the stool pigeonsI'ma do it for my people, bars are lethal, dawg, I'm running youThat was all from number one! Damn, who's number two?Black Thought from The Roots, I'm about to blast offYou know that I'm 'bout to do it like Future, my mask offToo sick with this, I'm risin' aboveNumber two is Black Thought, he's on a quest for loveHe's firm with The Roots, forever do my thing for the artYou steppin' incorrect to him, that's when things fall apartThat's a Roots album for y'all, you know I got 'em tremblin'Happy birthday, number three, y'all know we give it to EminemPuttin' it down for my people whenever I'm on itThey call me a true godNext dude ain't quite as lyrical, but love to Snoop DoggStevie, are you down with Snoop? You know he's got the flavorI know he ain't intricate with bars, but dawg, his vibes is majorAight, number five is transcendent with barsThe young legend outta Compton known as Kendrick LamarWe heard him flex on Good Kid, then flex againCan't believe I did a freestyle next to himNow, I know what you thinkin', "Okay, Harry, it's just a freestyle......At your level, you could just do that when you senile......We know what you do, you kick all them flows......You point out the color of a shirt, or some other clothes..."Nah, let me tell y'all what the fuck is really goin' onI ain't just rappin' 'bout T-shirts, dawg, I'm really flowin' strongSee, compared to rappers, I be all in they dreams'Cause I can take one singular word, then craft out a scheme"But Harry, wait, what's a scheme?" I'ma break that shit downThat'd be like if you gave me one word, let's say it's "town"Now I could do a simple word about the town that I'm fromOr go from town to city to building and stay on my runFrom building to foundation to skyscraper, they know I'm quite major...Ah, fuck, man!Ay, y'all can tap in on the channel @Stevie Knight later!Let's go![ Transcribed and subbed by Nick @plebcrabslayer ]Seventh Clip - Harry Potter, Octopus, JacketBirthin' the verse all up off the top o' the mind, all these lyrics get spawnedI should change my name to Harry Potter, I should change my mic to a wandYeah, I get loose with the tacticsUnh, and I'm tearin' through the fabricYeah, I'm really 'bout the raw rap shitI'm like Harry Potter 'cause I'm workin' magicUnh, and you know that I be hotter than hellEvery time I'm in the session, I be castin' a spellUnh, y'all know that I'm ready to slamI'm about to do a trick like, “Alakazam!”Unh, when I be rappin', I never be small-time, I be on some big shitAy, hop on my broomstick, now I be playin' some quidditchUnh, all on the field, ay, that's where you probably find meAy, ay, if I ain't there, then I'm with Ron, I'm with HermioneYeah, yeah, every time that I rhyme off the topI be playin' them god sportsAy, I'm workin' magic, yes, I am a wizard and I am at HogwartsUnh, every time I grip up on the mic, man, you know that I'm truly controllin'Ay, everything I write 'bout to make me a billionaire, that's J.K. RowlingUnh, and y'all know that she really been hittin'Harry Mack, off top when I'm spittin'Shouts out to J.K. RowlingShe said J.K. but she really wasn't kiddin'Really takin' over, built the empireWhen it come to lyrics, man, they know I been fireComin' off the tip o' this, I'ma get you high like lots o' kushI be goin' deep like a octopusIf it ain't natural, you gotsta push, I'ma get it doneEvery time that Harry Mack is freein' in a sessionI got everybody on Omegle havin' funLet's go, ay, spit what I thinkDon't need a squid, I don't use inkUnh, when I be on it, I'm spittin' the lyrics they sure to remember, thoughEverything that I spit gonna stick like a tentacleI put it deep in your mental, thoughHarry Mack, when it be comin' to lyricsI'm healin' your pain like the medicalMan, I got several styles I can use, I be up in my zoneI see you on the side, you filmin' on your phoneAnytime I grip the mic, you know I'm 'bout to damage youShouts out to your octopus, I'm diggin' that stuffed animalComin' off the tip o' this whether high or soberI said, "What you doin'?"They said, "It's a sleepover!"Mack goin' in, mad style, I pack itBrrr, too cold, you are gonna need a jacketH-Mack, I be flowin' off the headShout to the two fans layin' in the bedEvery time I'm on it, man, I really take controlI'm known around the globe, she in the white globeIn the white robe, that's what I meant to sayAny rapper who be steppin' is gettin' sent awayIt's Mack, I'm all up on it, and I gotta keep it honestEvery time I grab the mic, you know I keep it like a promiseGo beyond it when I'm on it with the double-time spittin'I be gettin' loose, no beat, still rhythmH-Mack, y'all know I really came to spit the flowPeace to the homie with the iPhone videoWhat!Last Clip - Guitar, Star Wars, Black Holeyeah all right let's get ituh check out the way that we sparkedthis hmac collab with marcuswe be locked in here on amigo weconnecting with brand new peopleyeah i'ma come off tip everybody tripwhenever i ripyeah can i keep it firm in my grip holdon the mic and it's never gonna slipyeah y'all know what come with the barswhen i be rhyming i'm traveling farhey[Music]but we don't need no guitar we gotmarcus all on the chords yeah he's onthe north it be the stage three in therays g i just come off tip if i said itthen i meant it if freestyle with guitari'd be hendrixi'm about to lay it down when they gotthe lyrics at the top of the middleeverybody knowing i don't really playaround let's gohey give you more y'all know i'm bout todrop bars heydon't play guitar but still i feel kindof like a rock star pop up on the screenand they know where it's atyeah every mag we become minor right nowfrom the bottom and you know i'm reallyclimbing right nowreal lyrics i know that i wrap bars i'llbe massive kind of like the death staryeah i've been coming through withcreations that's not a moonit's a space station h man i'm facingthe challenge leipzig kind of likemillennium falcon henry mack you knowthat i got the bomb vocals hit me like arookie i roll with han solo call mechewbacca you all know i'm about to betaller than a skyscraperhold up matter of fact i be a jedi[Music]i'm about to hit bad in my x-wing wheni'm on the mic i'll be flexing they'd bechecking they'd be watching the way thati'm moving always being proven boy sosoothingyeah i'm about to blast my enemies tofractalsevery time i rhyme i'm navigatingthrough a black holehey call him his majesty hey i use myfacultieshey kinda like black hole i hold it

Gengar the hard taco, Weedle the soft taco, and Alakazam the grilled cheese. Check out projectwarship.org to support Ukraine. --- Support this podcast: https://anchor.fm/threemicz/support

Alakazam Magic have been promising a brand new website for over a year now! There have been a few delays but now the website is actually here! In this interview with Harry & Peter Nardi, Craig finds out all about the new website. Peter and Harry talk all about everything that the website will make possible and how it is going to provide a much better experience for their customers. As an added bonus, Harry and Peter talk about new products which are coming soon to Alakazam Magic.

Stevo Watson has become a household name among magicians in the UK over the last few years!  He joined Ace Magic Studios a few years ago and quickly became known as an incredible creator and maker of gimmicks and gaffed cards.  He also started creating his own tricks and effects and over the last few years has created some amazing effects along with his partner Gary Sumpter. Then in 2021 Ace Magic Studios became part of Alakazam Magic and Stevo joined the Alakazam Magic team.  His reputation has gone from strength to strength and he is now a key part of the success of Alakazam Magic.

Yeah Wanna know what it's like (like) Baby, show me what it's like (like) I don't really got no type (type) I just wanna fuck all night Yeah-yeah, oh-whoa-whoa (oh, ooh, mmm) Baby, I need to know, mmm (yeah, need to know) I just been fantasizin' (size) And we got a lotta time (time) Baby, come throw the pipe (pipe) Gotta know what it's like (like) Yeah-yeah, oh-whoa-whoa Baby, I need to know, mmm What's your size? (Size) Add, subtract, divide ('vide) Daddy don't throw no curves (curves) Hold up, I'm goin' wide (wide) We could just start at ten (ten) Then we can go to five (five) I don't play with my pen (pen) I mean what I write Yeah-yeah, whoa-whoa-whoa I just can't help but be sexual (whoa) Tell me your schedule (yeah) I got a lotta new tricks for you, baby Just sayin' I'm flexible (I will) I do what I can to get you off (I will) Might just fuck him with my makeup on (I will) Eat it like I need an apron on (yeah, ay) Eat it 'til I need to change my thong (yeah, ay) We could do it to your favorite song (yeah, ay) Take a ride into the danger zone You know my nigga be buggin' me I just be wonderin' if you can fuck on me better Itchin' for me like an ugly sweater Need it in me like a Chuck E. need cheddar I need to know (yeah) Wanna know what it's like (like) Baby, show me what it's like (like) I don't really got no type (type) I just wanna fuck all night Yeah-yeah, oh-whoa-whoa (oh, ooh, mmm) Baby, I need to know, mmm (yeah, need to know) I just been fantasizin' (size) And we got a lotta time (time) Baby, come throw the pipe (pipe) Gotta know what it's like (like) Yeah-yeah, oh-whoa-whoa Baby, I need to know, mmm You're exciting, boy, come find me Your eyes told me, "Girl, come ride me" Fuck that feeling both us fighting Could he try me? (Yeah) mmm, most likely Tryna see if you could handle this ass Prolly give his ass a panic attack Sorry if I gave a random erection Prolly thinkin' I'm a telekinetic Oh, wait, you a fan of the magic? Poof, pussy like an Alakazam (yeah) I heard from a friend of a friend That that dick was a ten out of ten I can't stand it, just one night me Clink with the drink, gimme a sip Tell me what's your kink, gimme the dick Spank me, slap me, choke me, bite me (ew) Uh, wait, I can take it (ah) Give a fuck 'bout what your wifey's sayin' (yeah) Wanna know what it's like (like) Baby, show me what it's like (like) I don't really got no type (type) I just wanna fuck all night Yeah-yeah, oh-whoa-whoa (oh, ooh, mmm) Baby, I need to know, mmm (yeah, need to know) I just been fantasizin' (size) And we got a lotta time (time) Baby, come throw the pipe (pipe) Gotta know what it's like (like) Yeah-yeah, oh-whoa-whoa Baby, I need to know, mmm

*** Zaahruq breaks down simple magick to prepare for the Spring ***Yes, Indeed - Spring is coming.Are y'all ready for the seasonal pivot?

Joining Harlan this week is actress, writer and co-host of the Alakazam podcast, Samantha Brooks. The two chop it up about their inner fat kid habits, Gotham Magazine's top 49 NYC restaurant list and much, much more on this week's episode of The Sinatra Dinner Hour. --- This episode is sponsored by · Anchor: The easiest way to make a podcast. https://anchor.fm/app

Ever wanted to listen to an episode with technical difficulties? Now you can! In this special glitch edition of Alakazam with Elias and Sam, hear Samantha and half of Alex interview filmmaker, Elias Plagianos who talks working with Samantha on the set of SuperUser Do and writing for and directing Alakazam favorite Richard Kind in the Canneseries select pilot, Hudson Falls. Plus hear about Elias' creative process and his exciting new project.

The Ringer's Steve Ceruti of the Ryen Russillo Podcast joins the boys to share a few laughs and talk NBA basketball.Follow Ceruti - @ceruti Follow us @greenenvypod and @The617_

On this episode of The Sinatra Dinner Hour, Harlan is joined by the hosts of his new favorite podcast, Alakazam with Alex & Sam as they talk about their jump into the world of podcasting, odd and unusual snacks and of course their love of food! wee.thesinatradinnerhour.com --- This episode is sponsored by · Anchor: The easiest way to make a podcast. https://anchor.fm/app

In this episode we both have a close call. Scotty P. and Chadwick head to Mt. Chimney to take down a dog fighting ring. We then bunny hop to our fourth gym battle where things get a little heated. Scotty P. Finally gets his Alakazam, and Chadwick cops the new Pokémon Oreos. All of this and more on this weeks episode. --- This episode is sponsored by · Anchor: The easiest way to make a podcast. https://anchor.fm/app

Release boxes ready to open and the custom booster box is this Friday as well! --- This episode is sponsored by · Anchor: The easiest way to make a podcast. https://anchor.fm/app Support this podcast: https://anchor.fm/stalkgamers/support

The Daycare Dittos are beyond excited to welcome Steven, the literal brother of the podcast to his first episode! Join us for a tale of magicians, the occult, and spoon-based lawsuits as we unpack the psychic powerhouse Abra!

With MY word, i CREATE. AbRa-CaDaBra this is how i KNOW we have already W.O.N (wholeness, oneness, nowness) This Podcast Episode is honestly a vibe for YOU my TRiBE ! Mark England is & has been a teacher for years with a vision & PRO-active practice UNLOCKiNG the BREATH of HUMANiTY! His community & TRiBE are most known as the LANGUAGE & the BREATH people. First OFF it is an iNSiDE effort & practice to LEARN how we CREATE what we ALLOW our SELF to see. #ENLiFTED beyond the external by more FOCUS on the iNTERNAL, by transforming our inner-dialogue with something as SUBTLE as the practice known as the SOFT-TALK-CHALLENGE. Please DO iT for YO SELF !!! i know i am extremely grateful for this DiViNE connection with Mark & ALL of YOU amazing iNFiNiTE iNTER-DiMENSiONAL LiGHT BEiNGS

Abracadabra! Alakazam! Bippity boppity boopity!How do you live your best possible Life as a Wizard, you ask? Great question: your two favorite morons, Dangerous Boys, Wine Moms, Masters of Romance, and Pope Detectives Jesse McGrath & Wilder Shaw are here to answer that for you, in the form of an easy-to-digest, incredibly boring rubric. Tune into the show critics have hailed as "a podcast" to find out the Very Quite Lit, Very Pretty Good, and Very Not Quite Lit elements of life as a wizard!POINTS OF INTEREST THIS WEEK INCLUDE:• Moonbeam• Clam juice• Magic tears• Friends on the other side[Holler atcha boys on Instagram and Twitter]See Privacy Policy at https://art19.com/privacy and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info.

Transcript: https://bit.ly/3u6Acou Pay us. ko-fi.com/rateeverymon It is said to have an IQ of approximately 5,000. Its overflowing psychokinetic powers cause headaches to anyone nearby.

Host: Dan Holzman Guests: Al Millar talks about juggling, street performing, and his career as Alakazam.          

After a run in with Rocko, Filbert, and a mysterious stranger and his Alakazam, our heroes were able to rescue Natasha and Johanna from a contentious hostage situation! But now, Sen has his first gym battle just around the corner! Tune in to see what all the "buzz" is about!

Hello PUCLonians and Happy New Year! When you read this, it will be 2017! For a lot of people, 2016 wasn't a very good year for a variety of reasons. There are a myriad of reasons that 2016 was disappointing, but I won't go into those. This article is about celebrating what 2016 meant for Pokemon. 2016 was Pokemon's 20th Anniversary, and what an anniversary it was! The Pokemon craze hasn't been this big since the beginning of the franchise (if not bigger) and the past year has been the best time to be a Pokemon Fan!Games and ConsolesThis year was amazing for games. First, early in the year, the original Pokemon games came to the virtual console on 3DS and many people were able to enjoy the games that began the franchise. For some it may have been the first time to play these, and for others, it was the game that started it all for them. Not only were these games released, it was also made known that you would be able to transfer Pokemon in these games to PokeBank and … well, we will get to that.In March, Pokken Tournament, a fighting game from the makers of the Tekken series (my favorite fighting game series). Some people hopped on the band wagon for this right away, others stayed on and were excited to play this as an official event at Pokemon Championships. While this game may not have the staying power of the other Pokemon games, it is still an interesting addition to the franchise.Oh yeah! And some small games called Sun and Moon were announced and came out… All kidding aside, these games are a great step forward in the series. The games have been fun to play (even though I am playing slowly) and are arguably the best Pokemon games to date. At this time, I won't say anything else about these games, because so much has been said about them recently, but these games were the best way to cap off this amazing year of Pokemon!Also, each month this year (excluding January), there were mythical Pokemon released to XY and ORAS. This was the first time many of these Pokemon have been available since Gen III or IV, so it was pretty nice to be able to have access to these Pokemon again!MerchandiseThis year has seen some amazing merchandise! First, the varierty of toys, plushes (official and unofficial), along with shirts, mugs, pants, lanyards, and all sorts of things is astounding! Check out my Christmas article to see a lot of these items! But more than the regular items you would typically see, the 20th anniversary items have been awesome. Seen above are most of the 20th Anniversary figures that were released. And below is my personal collection of the mythical plushes.The figures, plushes, and TCG boxes (which I will cover next) corresponded with digital releases the mythical Pokemon.TCGNow, TCG isn't particularly my specialty, but this was an amazing year for the TCG, as well. As mentioned before, the Mythical collection was released giving promo cards for the Mythical Pokemon, as well as pins. Included in these were the Generations TCG boosters, which was one of many love letters that the TCG wrote this year. The other ones that stuck out to me were Fates Collide and Evolutions. Fates Collide was amazing to me because it basically featured my favorite Pokemon, Alakazam and had 5 unique cards for it! It was sweet! This has even inspired me to collect all of the Alakazam cards (as well as Abra and Kadabra) that have been printed. Evolutions stuck out to me because it was a reprint (kind of) of the original base set. I loved that set and have a collection of the entire set! It was great to see the art work again and see a few additions to the collection like Mega-Venusaur, Mega-Charizard, and Mega-Blastoise, among others!AnimeI can't say that I am the biggest fan of the anime, but I have watched a lot of the classic episodes with Snagorina and Snagorino, but this year was something special for the anime. The XYZ saga in the anime has gotten much acclaim for not just being another Saturday morning cartoon, but a true anime that many people enjoy. Gamexplain did an entire 45 minute episode discussing the virtues of this show. Ash had a really good team and even made it to the Kalos league finals!While I may not have focused much on the traditional anime, Pokemon Generations was something that I made sure I watched every Friday! These mini episodes released on YouTube featured many scenes that the player character wasn't able to see during the play through of the game. These had a lot of emphasis on the rivals in the games, the evil teams, as well as Looker. It was an amazing set of episodes (for the most part) that I really appreciate and am glad were included this year. Check out Ethan and Jushiro reviewing the first 6 episode on PUCL Movie Club!Pokemon GoPokemon Go might be the biggest Pokemon story this year! This game swept the country and the world for a few months this summer. It was the hottest thing out there. Scores of people were out walking and looking for Pokemon, many of whom had never played the series before. It was amazing to see how many people actually got involved and have become fans because of it. While the hype may have died down a bit since the initial push, the game is still incredibly popular and was one of the big reasons Pokemon was so popular this year!PUCLWhile PUCL may have only been a small contribution to what the Pokemon 20th anniversary was, 2016 was an amazing year for PUCL. This year PUCL was able to start a Patreon which was able to help us support our community more and provide some really cool things to our Patrons, as well as the community as a whole. The TCGcast didn't start this year, but came into full swing and even expanded to become a bi-monthly show! The Movie Club started this year, as well as the recently rebranded Koffee Hour (formerly known as PUCL L.I.V.E) and at the very end of the year we added Game Corner and the Battle Cast. We also added a few writers to the group such as Antonin and Capt. Gravy who have been a great contribution to the writing staff. We have so much content coming out right now!!!And I couldn't talk about 2016 without talking about PUCLcon. We had a score of people descend upon Thatch's house from all parts of the country, continent, and even the world! It was an amazing event, not only because of the Nationals event that was going on, but mostly because we had the opportunity to meet people you had maybe only heard on the podcast, or had only met in the chatroom. And we had 4 days to just hang out and play Pokemon. It was a fantastic time and I am definitely looking forward to going to PUCLcon in 2017, where ever it may be.One more comment I need to make about 2016 and that is that 2016 was the year of R. Sigma. While R. Sigma did not participate in the UUTC at the start of the year because he was focusing on double, he won basically everything after that. He is the PFTT Champion, Summer League Champion, and Generations Tournament Champion. In addition to that, he also won the first unofficial PUCL TCG Tournament, as well as the Pokemon Underground Fantasy Football League (PUFFL). R. Sigma, you are a fantastic competitor and alway humble in your victories! I can safely say that you are the PUCL Champion for 2016. Great job!With that PUCLonians, I must sign off. But I did want to take this time to remind you that while 2016 might have sucked in many aspects for some people, for Pokemon, 2016 was the best of times. Don't forget this time, friends, because it might be another 10 years before we see a celebration quite like that one we had this year.Have a wonderful 2017 everyone!~Prof. SnagMake a writer's day. Leave a comment! Hosted on Acast. See acast.com/privacy for more information.