|
67843
|
My awesome macropad! by @Headhunter | Stardance
|
Headhunter
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My awesome macropad! by @Headhunter | Stard Title: My awesome macropad! by @Headhunter | Stardance...
|
1793735468.0171995
|
1794340268.0171995
|
|
67842
|
NULL
|
NULL
|
0
|
0.0
|
0
|
[]
|
[]
|
|
1793760652.9876878
|
1794624652.9876878
|
|
67841
|
NULL
|
NULL
|
0
|
0.0
|
0
|
[]
|
[]
|
|
1793760649.1199853
|
1794624649.1199853
|
|
67840
|
NULL
|
NULL
|
0
|
0.0
|
0
|
[]
|
[]
|
|
1793760647.416916
|
1794624647.416916
|
|
67839
|
Amanyu by @amanyu | Stardance
|
amanyu
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Amanyu by @amanyu | Stardance
|
1793735465.9685564
|
1794340265.9685564
|
|
67838
|
SafeDispose by @spa | Stardance
|
spa
|
1
|
11.0
|
0
|
[{"ship_num": 1, "ts": "2026-1 [{"ship_num": 1, "ts": "2026-10-24T19:16:23Z", "hours": 11.0, "multiplier": null, "payout": 0, "blessing": "neutral", "text": "I made SafeDispose, a household disposal guide for Orange County residents! Users can search or browse through many different categories of items to find disposal instructions, safety guidance, and official sources. They can also save items to a disposal list, which matches their saved items to disposal collection locations that accept them, with directions and official location information. One of the biggest challenges I faced was keeping all of the information and items organized as I added more categories. I combated this by giving each and every item a unique ID, which made it much easier to organize all the code and access the items whenever needed. I am especially proud of how well the disposal list and the location-matching system turned out! To try it out, search for an item using the search bar or browse the categories, add a few items to your disposal list, and open the list to see matching locations and disposal options! You can also view more information on my project on the homepage.", "repo": "https://github.com/spancharapula-byte/SafeDispose", "demo": "https://safedispose.netlify.app/"}]...
|
[{"post_id": 82414, "ts": "202 [{"post_id": 82414, "ts": "2026-10-24T19:01:45Z", "hours": 0.375, "text": "I added a Show More/Show Less button for the disposal locations to keep it less cluttered and more organized. I also added a new section for saved items that don’t have matching drop-off locations, so users can still access their disposal instructions even when no matching locations are listed. I also made some final UI adjustments and tested all the features to make sure everything works properly."}, {"post_id": 80065, "ts": "2026-10-16T04:06:56Z", "hours": 1.0869, "text": "I redesigned and cleaned up the overall UI, including buttons, search results, safety notes, and disposal location cards and gave it a much cleaner look. I also updated the homepage by adding a How it works section, About section, and a footer."}, {"post_id": 77474, "ts": "2026-10-07T04:40:32Z", "hours": 0.6328, "text": "I finished adding disposal information for the remaining categories, including Medicine & Needles, Automotive, Appliances, and Furniture & Bulky Items. I also went through and reviewed all the disposal information for every category for consistency and to fix any typos or grammatical errors."}, {"post_id": 77227, "ts": "2026-10-06T05:35:41Z", "hours": 0.6411, "text": "Cleaned up inconsistencies for disposal information across several items and categories. I fixed up inconsistent wording for hazardous waste locations, safety instructions, and corrected a few disposal details to keep consistency across similar items."}, {"post_id": 76619, "ts": "2026-10-04T06:07:23Z", "hours": 8.1, "text": "SafeDispose is an Orange County disposal guide I’ve been building to help people easily figure out how to safely dispose of their everyday items. So far, I’ve added multiple categories and sub-categories of items, a search feature, specific disposal instructions backed by sources for each item, a disposal list you can save items to, and a system that matches saved items to Orange County household hazardous waste drop-off locations. I also added Google Maps directions links for the hazardous waste center locations and official source links."}]...
|
Title: SafeDispose by @spa | Stardance
Devlogs:
S Title: SafeDispose by @spa | Stardance
Devlogs:
SafeDispose is an Orange County disposal guide I’ve been building to help people easily figure out how to safely dispose of their everyday items. So far, I’ve added multiple categories and sub-categories of items, a search feature, specific disposal instructions backed by sources for each item, a disposal list you can save items to, and a system that matches saved items to Orange County household hazardous waste drop-off locations. I also added Google Maps directions links for the hazardous waste center locations and official source links.
Cleaned up inconsistencies for disposal information across several items and categories. I fixed up inconsistent wording for hazardous waste locations, safety instructions, and corrected a few disposal details to keep consistency across similar items.
I finished adding disposal information for the remaining categories, including Medicine & Needles, Automotive, Appliances, and Furniture & Bulky Items. I also went through and reviewed all the disposal information for every category for consistency and to fix any typos or grammatical errors.
I redesigned and cleaned up the overall UI, including buttons, search results, safety notes, and disposal location cards and gave it a much cleaner look. I also updated the homepage by adding a How it works section, About section, and a footer.
I added a Show More/Show Less button for the disposal locations to keep it less cluttered and more organized. I also added a new section for saved items that don’t have matching drop-off locations, so users can still access their disposal instructions even when no matching locations are listed. I also made some final UI adjustments and tested all the features to make sure everything works properly.
Ships:
I made SafeDispose, a household disposal guide for Orange County residents! Users can search or browse through many different categories of items to find disposal instructions, safety guidance, and official sources. They can also save items to a disposal list, which matches their saved items to disposal collection locations that accept them, with directions and official location information. One of the biggest challenges I faced was keeping all of the information and items organized as I added more categories. I combated this by giving each and every item a unique ID, which made it much easier to organize all the code and access the items whenever needed. I am especially proud of how well the disposal list and the location-matching system turned out! To try it out, search for an item using the search bar or browse the categories, add a few items to your disposal list, and open the list to see matching locations and disposal options! You can also view more information on my project on the homepage....
|
1793930046.0631707
|
1794016446.0631707
|
|
67837
|
My Cool blogs by @prathamrawat | Stardance
|
prathamrawat
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My Cool blogs by @prathamrawat | Stardance
|
1793735459.346408
|
1794340259.346408
|
|
67836
|
NULL
|
NULL
|
0
|
0.0
|
0
|
[]
|
[]
|
|
1793760646.562166
|
1794624646.562166
|
|
67835
|
GDEV by @gdev | Stardance
|
gdev
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: GDEV by @gdev | Stardance
|
1793458593.6821535
|
1794322593.6821535
|
|
67834
|
VerityOS by @shogunxfortnite | Stardance
|
shogunxfortnite
|
1
|
1.0
|
0
|
[{"ship_num": 1, "ts": "2026-1 [{"ship_num": 1, "ts": "2026-10-05T08:13:28Z", "hours": 1.0, "multiplier": null, "payout": 0, "blessing": "neutral", "text": "I made a custom OS called VerityOS.", "repo": "https://github.com/shogunxfortnite-cyber/VerityOS", "demo": "https://verityos2.netlify.app/"}]...
|
[{"post_id": 76984, "ts": "202 [{"post_id": 76984, "ts": "2026-10-05T08:09:45Z", "hours": 0.6269, "text": "I am working on a VerityOS for the challenge webos 1. I am done with it you can try it in my git repo"}]...
|
Title: VerityOS by @shogunxfortnite | Stardance
D Title: VerityOS by @shogunxfortnite | Stardance
Devlogs:
I am working on a VerityOS for the challenge webos 1. I am done with it you can try it in my git repo
Ships:
I made a custom OS called VerityOS....
|
1793839803.8727288
|
1793926203.8727288
|
|
67833
|
NULL
|
NULL
|
0
|
0.0
|
0
|
[]
|
[]
|
|
1793760645.703698
|
1794624645.703698
|
|
67832
|
autolink by @AmAlan46 | Stardance
|
AmAlan46
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: autolink by @AmAlan46 | Stardance
|
1793735456.3526988
|
1794340256.3526988
|
|
67831
|
Geeked by @TheGeekachu | Stardance
|
TheGeekachu
|
1
|
3.0
|
0
|
[{"ship_num": 1, "ts": "2026-1 [{"ship_num": 1, "ts": "2026-10-24T17:57:40Z", "hours": 3.0, "multiplier": null, "payout": 0, "blessing": "neutral", "text": "I made Geeked, a blog for geeks and engineering enthusiasts! Thanks for being here, reading my devlogs, and putting up with my terrible coding skills! Implemented with Better Auth and MongoDB, using the same CSS styling as my personal site. I had to do a bit of unorthodox stuff, but overall, I like this blog! Enjoy!", "repo": "https://github.com/TheGeekachu/geeked-blog", "demo": "https://geeked.thegeekachu.dev"}]...
|
[{"post_id": 77165, "ts": "202 [{"post_id": 77165, "ts": "2026-10-05T21:06:37Z", "hours": 1.0158, "text": "I did a bunch of security stuff. I know it says 1 hour but it’s more like 3, because of Vercel and Better Auth being jerks. Anyway, Vercel and Better Auth and MongoDB weren’t playing nice with each other, each saying the others had unsafe settings and refusing to talk to my deployed app, problems I didn’t have on my dev server. So, with a lot of Gemini, I finally fixed it, and got a Better Auth dashboard up and running!"}, {"post_id": 76844, "ts": "2026-10-04T20:04:40Z", "hours": 1.9433, "text": "I finished most of th blog, some final bugs exist, but apart from an unwritten README, we’re ready to ship! Re-used CSS from my personal site."}]...
|
Title: Geeked by @TheGeekachu | Stardance
Devlogs Title: Geeked by @TheGeekachu | Stardance
Devlogs:
I finished most of th blog, some final bugs exist, but apart from an unwritten README, we’re ready to ship! Re-used CSS from my personal site.
I did a bunch of security stuff. I know it says 1 hour but it’s more like 3, because of Vercel and Better Auth being jerks. Anyway, Vercel and Better Auth and MongoDB weren’t playing nice with each other, each saying the others had unsafe settings and refusing to talk to my deployed app, problems I didn’t have on my dev server. So, with a lot of Gemini, I finally fixed it, and got a Better Auth dashboard up and running!
Ships:
I made Geeked, a blog for geeks and engineering enthusiasts! Thanks for being here, reading my devlogs, and putting up with my terrible coding skills! Implemented with Better Auth and MongoDB, using the same CSS styling as my personal site. I had to do a bit of unorthodox stuff, but overall, I like this blog! Enjoy!...
|
1793929989.8958561
|
1794016389.8958561
|
|
67830
|
My awesome macropad! by @Dinakar | Stardance
|
Dinakar
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My awesome macropad! by @Dinakar | Stardanc Title: My awesome macropad! by @Dinakar | Stardance...
|
1793735452.1441584
|
1794340252.1441584
|
|
67829
|
TrackPoint-style mouse for tight spaces by @smitam TrackPoint-style mouse for tight spaces by @smitamaw | Stardance...
|
smitamaw
|
0
|
1.33
|
0
|
[]
|
[{"post_id": 78218, "ts": "202 [{"post_id": 78218, "ts": "2026-10-10T00:34:16Z", "hours": 0.8836, "text": "Devlog #2 I worked a little more on my CAD model for my TrackPoint-style mouse device. This device is intended to fit in tight spaces where a traditional mouse wouldn’t work, such as if you are using a laptop on your lap or on a very small desk. In my CAD model, I added a space/slot for the ESP32S3 that processes imputs from the joystick and buttons and sends mouse movements and actions to the computer via USB cable. (I use a USB-C to USB-A male-to-male cable to attach the device to the computer.) Next up: printing out the case and taping all the components in place!"}, {"post_id": 76613, "ts": "2026-10-04T05:34:04Z", "hours": 0.45, "text": "I just wired up the first prototype of my new project–a mouse that uses a joystick and buttons just like Lenovo ThinkPad TrackPoint, for use in spaces where a traditional mouse would not have enough freedom of movement. The reason I am building this is because, in most of my classes in high school, a laptop is required. However, I don’t want to use the trackpad, and the mouse that I bring usually doesn’t get enough space to move around on my small desk. So I am creating this project to help me use my laptop better and more conveniently at school. In addition to wiring up my first prototype, I also CADded up a first prototype for the 3D-printed frame the components are going to rest on. (It isn’t going to last long if it stays taped onto a folded index card!) Hopefully in the next devlog I will have printed the frame and maybe even mounted the parts to it!"}]...
|
Title: TrackPoint-style mouse for tight spaces by Title: TrackPoint-style mouse for tight spaces by @smitamaw | Stardance
Devlogs:
I just wired up the first prototype of my new project–a mouse that uses a joystick and buttons just like Lenovo ThinkPad TrackPoint, for use in spaces where a traditional mouse would not have enough freedom of movement. The reason I am building this is because, in most of my classes in high school, a laptop is required. However, I don’t want to use the trackpad, and the mouse that I bring usually doesn’t get enough space to move around on my small desk. So I am creating this project to help me use my laptop better and more conveniently at school. In addition to wiring up my first prototype, I also CADded up a first prototype for the 3D-printed frame the components are going to rest on. (It isn’t going to last long if it stays taped onto a folded index card!) Hopefully in the next devlog I will have printed the frame and maybe even mounted the parts to it!
Devlog #2 I worked a little more on my CAD model for my TrackPoint-style mouse device. This device is intended to fit in tight spaces where a traditional mouse wouldn’t work, such as if you are using a laptop on your lap or on a very small desk. In my CAD model, I added a space/slot for the ESP32S3 that processes imputs from the joystick and buttons and sends mouse movements and actions to the computer via USB cable. (I use a USB-C to USB-A male-to-male cable to attach the device to the computer.) Next up: printing out the case and taping all the components in place!...
|
1793735448.8469765
|
1794340248.8469765
|
|
67828
|
Arc Study Tracker by @flamex | Stardance
|
flamex
|
1
|
29.0
|
0
|
[{"ship_num": 1, "ts": "2026-1 [{"ship_num": 1, "ts": "2026-10-04T05:16:41Z", "hours": 29.0, "multiplier": null, "payout": 0, "blessing": "neutral", "text": "i built this small corner of mine to track my study hours and hopefully plan the overall academic part of my life. it features different meaningful tools for better productivity like session logging through a timer, detailed insights of your sessions, and grade tracking. This is my first time making such a webapp so i had to use some help in AI but im really proud of what the result turned out to be. p.s. : I’ve been using this tracker regularly for tracking my progress The Database for the project is actually unified so if you track the hours in the website rn , it’ll be added to the main database of mine. I wanted to add multi user support but thats something im currently learning about and i didn’t have much idea of how to make the whole flow so next time ill make a web-app through react , ill add the multi user thing too and hopefully a fullstack project. YAYYYYYYYY LESGOOOOO", "repo": "https://github.com/Diparsan79/arc", "demo": "https://arc.fl4me.tech/"}]...
|
[{"post_id": 76610, "ts": "202 [{"post_id": 76610, "ts": "2026-10-04T05:11:29Z", "hours": 0.3003, "text": "Finally the working website is COMPLETE….. i built this small corner of mine to track my study hours and hopefully plan the overall academic part of my life. it features different meaningful tools for better productivity like session logging through a timer, detailed insights of your sessions, and grade tracking. After making LIT a terminal based python app , i wanted to learn more deeply by making an full stack app. Thats when i stumbled upon the idea for arc. I have tried many different kinds of study tracker apps but i wanted to make on of my own with my own kinda principles so i made THISSSSS. for some ai declarances I used ai as a programming mentor helping in different aspects of the site’s developmentai was used to help me structure some of the frontend components and some database structuring"}, {"post_id": 76604, "ts": "2026-10-04T04:47:09Z", "hours": 29.0542, "text": "I did fully thorough planning for the site with hand written features and brainstorms.(p.s. I actually did all this planning 2 or 3 days ago but i forgot to journal and share it here) i like to journal and document my projects but i forget about it. To sum up what ive done for the site till now: i made the timer html and js and im thinking of the log session taking you to the timer and then you can choose among different kinds of timings (pomodoro) or specific timers or just stopwatch and see where you’re going. im mostly going for the ipad aesthetics kinda thing where you start the timer and keep the ipad infront of you and you start studying. ( i plan on doing it after i buy the ipad through macondo after completing this and other projects) the social features are kinda hard for me but i think it’ll be quite challenging and tricky for me so i like that. I wanna add social layer where we can see our friends and so on but idk how that works as of now. so i gotta do a lott of research i think. HECK im still figuring out how the individual users will use it. im trynna get criticisms and learn from it. idk much javascript so i ask claude through sum gudd prompt engineering to review my js functionality that ive used and it helps me improve aswell"}]...
|
Title: Arc Study Tracker by @flamex | Stardance
D Title: Arc Study Tracker by @flamex | Stardance
Devlogs:
I did fully thorough planning for the site with hand written features and brainstorms.(p.s. I actually did all this planning 2 or 3 days ago but i forgot to journal and share it here) i like to journal and document my projects but i forget about it. To sum up what ive done for the site till now: i made the timer html and js and im thinking of the log session taking you to the timer and then you can choose among different kinds of timings (pomodoro) or specific timers or just stopwatch and see where you’re going. im mostly going for the ipad aesthetics kinda thing where you start the timer and keep the ipad infront of you and you start studying. ( i plan on doing it after i buy the ipad through macondo after completing this and other projects) the social features are kinda hard for me but i think it’ll be quite challenging and tricky for me so i like that. I wanna add social layer where we can see our friends and so on but idk how that works as of now. so i gotta do a lott of research i think. HECK im still figuring out how the individual users will use it. im trynna get criticisms and learn from it. idk much javascript so i ask claude through sum gudd prompt engineering to review my js functionality that ive used and it helps me improve aswell
Finally the working website is COMPLETE….. i built this small corner of mine to track my study hours and hopefully plan the overall academic part of my life. it features different meaningful tools for better productivity like session logging through a timer, detailed insights of your sessions, and grade tracking. After making LIT a terminal based python app , i wanted to learn more deeply by making an full stack app. Thats when i stumbled upon the idea for arc. I have tried many different kinds of study tracker apps but i wanted to make on of my own with my own kinda principles so i made THISSSSS. for some ai declarances I used ai as a programming mentor helping in different aspects of the site’s developmentai was used to help me structure some of the frontend components and some database structuring
Ships:
i built this small corner of mine to track my study hours and hopefully plan the overall academic part of my life. it features different meaningful tools for better productivity like session logging through a timer, detailed insights of your sessions, and grade tracking. This is my first time making such a webapp so i had to use some help in AI but im really proud of what the result turned out to be. p.s. : I’ve been using this tracker regularly for tracking my progress The Database for the project is actually unified so if you track the hours in the website rn , it’ll be added to the main database of mine. I wanted to add multi user support but thats something im currently learning about and i didn’t have much idea of how to make the whole flow so next time ill make a web-app through react , ill add the multi user thing too and hopefully a fullstack project. YAYYYYYYYY LESGOOOOO...
|
1793839801.9377973
|
1793926201.9377973
|
|
67827
|
Frictionless by @west | Stardance
|
west
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Frictionless by @west | Stardance
|
1793735446.7756288
|
1794340246.7756288
|
|
67826
|
Orion Desktop Pet by @KibbTheFox | Stardance
|
KibbTheFox
|
1
|
4.0
|
0
|
[{"ship_num": 1, "ts": "2026-1 [{"ship_num": 1, "ts": "2026-10-10T17:03:24Z", "hours": 4.0, "multiplier": null, "payout": 0, "blessing": "neutral", "text": "Desktop Pet Ship! Download here: Itch Github Github Repo I have an OC named Orion who is a cat so I thought it’d be fun to make a desktop pet of them for this ysws! Orion is 15 year old cat. She is very sweet, but is often stressed out by something. She likes thunder, winter, and tea, and thinks that white isn’t a color. Working with moving the window was a huge challenge, due to the mouse’s relative position to the window and also the fact that the mouse position is x2 compared to the window (i think ) I didn’t really follow the guide so I had to figure most of this out myself. (I hope its a unique project because I didn’t follow the guide) The art was made by me in procreate and pixlr. I did the outline on my ipad in procreate and colored it in pixlr later on into development. (I was unsure if I wanted to make them my fursona Kibb or Orion) I didn’t really track the time spent on art", "repo": "https://github.com/kibbthefox/Orion-Desktop-Pet", "demo": "https://corgtheporg.itch.io/orion-desktop-pet"}]...
|
[{"post_id": 78383, "ts": "202 [{"post_id": 78383, "ts": "2026-10-10T16:39:24Z", "hours": 0.7956, "text": "Desktop Pet Devlog 5: Release!!! Added fade out to pet counter Made the README Released the project I forgot how tweening worked so I had to reference the code from another project I made It was pretty easy with the reference tho. I also worked on exporting the project since I’m about done. README wasn’t too complicated to add, I just did what I’ve done before."}, {"post_id": 78254, "ts": "2026-10-10T04:56:18Z", "hours": 0.4989, "text": "Desktop Pet Devlog 4 Added petting/meowing w/ a counter Added new art I didn’t have much trouble implenting this. The main hurdle was just distinguishing when I was petting the cat and when I was grabbing it. I decided to game-ify it a little with the counter."}, {"post_id": 78249, "ts": "2026-10-10T04:00:16Z", "hours": 0.9711, "text": "Desktop Pet Devlog 3 Added dragging This was way hard than expected. I spent a while figuring out the equation to make the window to go to the mouse cursor, since the mouse cursor’s position was relative to the window. Eventually I figured that out, but then I had to figure out the cursor type because that broke in the process too. I got it all fixed tho! Hopefully I can start adding some better art next devlog."}, {"post_id": 77154, "ts": "2026-10-05T20:10:06Z", "hours": 0.7475, "text": "Desktop Pet Devlog 2 Added cursor changes/system Removed movement I got rid of the movement of the pet because I want to do some other system instead. I want to make it so you can click and drag them. I coded a simple version of this with only the cursor changing but then the mouse would jitter on the edges of the area you could grab so I had to write a little more complicated code to fix this. Next update I plan to add the grab and drag feature!"}, {"post_id": 76815, "ts": "2026-10-04T17:48:00Z", "hours": 0.8056, "text": "Desktop Pet Devlog 1 Created base sprite Adding following the cursor (tho it is broken) mouse position does not equal window position in measure or something so when my mouse gets to the edge of the screen, the window is still towards the middle."}]...
|
Title: Orion Desktop Pet by @KibbTheFox | Stardanc Title: Orion Desktop Pet by @KibbTheFox | Stardance
Devlogs:
Desktop Pet Devlog 1 Created base sprite Adding following the cursor (tho it is broken) mouse position does not equal window position in measure or something so when my mouse gets to the edge of the screen, the window is still towards the middle.
Desktop Pet Devlog 2 Added cursor changes/system Removed movement I got rid of the movement of the pet because I want to do some other system instead. I want to make it so you can click and drag them. I coded a simple version of this with only the cursor changing but then the mouse would jitter on the edges of the area you could grab so I had to write a little more complicated code to fix this. Next update I plan to add the grab and drag feature!
Desktop Pet Devlog 3 Added dragging This was way hard than expected. I spent a while figuring out the equation to make the window to go to the mouse cursor, since the mouse cursor’s position was relative to the window. Eventually I figured that out, but then I had to figure out the cursor type because that broke in the process too. I got it all fixed tho! Hopefully I can start adding some better art next devlog.
Desktop Pet Devlog 4 Added petting/meowing w/ a counter Added new art I didn’t have much trouble implenting this. The main hurdle was just distinguishing when I was petting the cat and when I was grabbing it. I decided to game-ify it a little with the counter.
Desktop Pet Devlog 5: Release!!! Added fade out to pet counter Made the README Released the project I forgot how tweening worked so I had to reference the code from another project I made It was pretty easy with the reference tho. I also worked on exporting the project since I’m about done. README wasn’t too complicated to add, I just did what I’ve done before.
Ships:
Desktop Pet Ship! Download here: Itch Github Github Repo I have an OC named Orion who is a cat so I thought it’d be fun to make a desktop pet of them for this ysws! Orion is 15 year old cat. She is very sweet, but is often stressed out by something. She likes thunder, winter, and tea, and thinks that white isn’t a color. Working with moving the window was a huge challenge, due to the mouse’s relative position to the window and also the fact that the mouse position is x2 compared to the window (i think ) I didn’t really follow the guide so I had to figure most of this out myself. (I hope its a unique project because I didn’t follow the guide) The art was made by me in procreate and pixlr. I did the outline on my ipad in procreate and colored it in pixlr later on into development. (I was unsure if I wanted to make them my fursona Kibb or Orion) I didn’t really track the time spent on art...
|
1793839799.0242312
|
1793926199.0242312
|
|
67825
|
TVC Rocket by @NoahH35725 | Stardance
|
NoahH35725
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: TVC Rocket by @NoahH35725 | Stardance
|
1793735444.5219095
|
1794340244.5219095
|
|
67824
|
My very own WebOS by @divypatel211020 | Stardance
|
divypatel211020
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My very own WebOS by @divypatel211020 | Sta Title: My very own WebOS by @divypatel211020 | Stardance...
|
1793735441.287824
|
1794340241.287824
|
|
67823
|
Untitled project by @pritamk980 | Stardance
|
pritamk980
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Untitled project by @pritamk980 | Stardance
|
1793735440.400159
|
1794340240.400159
|
|
67822
|
Tay's coding language efficient coding languag Tay's coding language efficient coding language to use on my ISS robot? by @taylan_abramson | Stardance...
|
taylan_abramson
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Tay's coding language efficient coding Title: Tay's coding language efficient coding language to use on my ISS robot? by @taylan_abramson | Stardance...
|
1793458580.6450455
|
1794322580.6450455
|
|
67821
|
WarioWare Game by @este558 | Stardance
|
este558
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: WarioWare Game by @este558 | Stardance
|
1793735439.4980843
|
1794340239.4980843
|
|
67820
|
Flappy Bird! by @soumilyadav25 | Stardance
|
soumilyadav25
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Flappy Bird! by @soumilyadav25 | Stardance
|
1793735437.4402564
|
1794340237.4402564
|
|
67819
|
joypad by @sunidhi1422 | Stardance
|
sunidhi1422
|
0
|
5.51
|
0
|
[]
|
[{"post_id": 77387, "ts": "202 [{"post_id": 77387, "ts": "2026-10-06T19:11:55Z", "hours": 5.5142, "text": "i am working on a esp8266 based small handy gamepad fidgit toy with several games in it"}]...
|
Title: joypad by @sunidhi1422 | Stardance
Devlogs Title: joypad by @sunidhi1422 | Stardance
Devlogs:
i am working on a esp8266 based small handy gamepad fidgit toy with several games in it...
|
1793735434.0887403
|
1794340234.0887403
|
|
67818
|
Autonomous Drone by @Maxwell_Li | Stardance
|
Maxwell_Li
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Autonomous Drone by @Maxwell_Li | Stardance
|
1793735426.509914
|
1794340226.509914
|
|
67817
|
Foldable Chess set by @Boyan | Stardance
|
Boyan
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Foldable Chess set by @Boyan | Stardance
|
1793735424.4404294
|
1794340224.4404294
|
|
67816
|
My very own WebOS by @pratik152008 | Stardance
|
pratik152008
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My very own WebOS by @pratik152008 | Starda Title: My very own WebOS by @pratik152008 | Stardance...
|
1793735422.4110389
|
1794340222.4110389
|
|
67815
|
My Cool Blog by @LOCKIN | Stardance
|
LOCKIN
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My Cool Blog by @LOCKIN | Stardance
|
1793735420.365693
|
1794340220.365693
|
|
67814
|
NULL
|
NULL
|
0
|
0.0
|
0
|
[]
|
[]
|
|
1793760643.97233
|
1794624643.97233
|
|
67813
|
NASA BLOG by @Lalith | Stardance
|
Lalith
|
1
|
10.87
|
0
|
[{"ship_num": 1, "ts": "2026-1 [{"ship_num": 1, "ts": "2026-10-07T19:01:08Z", "hours": 10.0, "multiplier": null, "payout": 0, "blessing": "neutral", "text": "Mission Log is a NASA-themed full-stack blog built with Next.js, React, Mongoose, and MongoDB, featuring Markdown articles, accounts, search, NASA data explorers, saved articles, and an editorial studio. The biggest challenges were connecting Atlas and Vercel, configuring environment variables, fixing author login, and restoring the redesigned site after an older deployment replaced it. The part I’m proudest of is how the original blog idea grew into a space publication where people can read stories, explore related data, and save what interests them. To test it, visit Mission Log, open articles, browse topics, try search, and explore the NASA data pages; account features include saving articles and following topics, while publishing requires authorization. Database credential recovery is still unfinished, so database-backed features may fail until configuration is updated. The project was developed with slight AI assistance, and Mission Log is not affiliated with NASA.", "repo": "https://github.com/Hustlenix/mission-log", "demo": "https://mission-log-omega.vercel.app/"}]...
|
[{"post_id": 79502, "ts": "202 [{"post_id": 79502, "ts": "2026-10-14T10:25:53Z", "hours": 0.8728, "text": "I wrote the whole read me by myself, ig it wil get shipped!"}, {"post_id": 77645, "ts": "2026-10-07T18:51:58Z", "hours": 0.7572, "text": "I’ve sucess fully finished this project, thats what i gi=uess. This was a good journey of that API keys and making a full stack website"}, {"post_id": 77280, "ts": "2026-10-06T12:38:11Z", "hours": 2.3531, "text": "NASA"}, {"post_id": 77092, "ts": "2026-10-05T16:54:38Z", "hours": 2.0228, "text": "It just looks like i have sorted everything out, and it is ready to be shipped, but this has few major problems like: the search isnt good the blog is not brodned to match a complete blog, this all feels kinda mid and not finished, so i set my AI to plan, research, and webscrape NASA and help me build a good blog site for NASA!!!"}, {"post_id": 76754, "ts": "2026-10-04T14:27:19Z", "hours": 1.4617, "text": "I have finished the overall UI/Ux of it and now i am on the mission to do those ENV files to make it a Full stack website, man why is it soo hard to make a full stack website man? i was well of when i was just useing my github pages for everyting😭😭, setting up the keys is way harder. Like creating a project, pushing the code, signing in for mongoDB or smtg. and also after doing all of these it will still say, deployment failed🙏🏻😭"}, {"post_id": 76594, "ts": "2026-10-04T03:12:52Z", "hours": 2.9228, "text": "I am on the mission to make a better blog than Nasa’s Blog, using their API key’s help and also webscrape them and use it. You can also get one from this link: https://api.nasa.gov/ that too for free"}]...
|
Title: NASA BLOG by @Lalith | Stardance
Devlogs:
Title: NASA BLOG by @Lalith | Stardance
Devlogs:
I am on the mission to make a better blog than Nasa’s Blog, using their API key’s help and also webscrape them and use it. You can also get one from this link: https://api.nasa.gov/ that too for free
I have finished the overall UI/Ux of it and now i am on the mission to do those ENV files to make it a Full stack website, man why is it soo hard to make a full stack website man? i was well of when i was just useing my github pages for everyting😭😭, setting up the keys is way harder. Like creating a project, pushing the code, signing in for mongoDB or smtg. and also after doing all of these it will still say, deployment failed🙏🏻😭
It just looks like i have sorted everything out, and it is ready to be shipped, but this has few major problems like: the search isnt good the blog is not brodned to match a complete blog, this all feels kinda mid and not finished, so i set my AI to plan, research, and webscrape NASA and help me build a good blog site for NASA!!!
NASA
I’ve sucess fully finished this project, thats what i gi=uess. This was a good journey of that API keys and making a full stack website
I wrote the whole read me by myself, ig it wil get shipped!
Ships:
Mission Log is a NASA-themed full-stack blog built with Next.js, React, Mongoose, and MongoDB, featuring Markdown articles, accounts, search, NASA data explorers, saved articles, and an editorial studio. The biggest challenges were connecting Atlas and Vercel, configuring environment variables, fixing author login, and restoring the redesigned site after an older deployment replaced it. The part I’m proudest of is how the original blog idea grew into a space publication where people can read stories, explore related data, and save what interests them. To test it, visit Mission Log, open articles, browse topics, try search, and explore the NASA data pages; account features include saving articles and following topics, while publishing requires authorization. Database credential recovery is still unfinished, so database-backed features may fail until configuration is updated. The project was developed with slight AI assistance, and Mission Log is not affiliated with NASA....
|
1793839795.2580993
|
1793926195.2580993
|
|
67812
|
WarioWare Game by @anul_liyanage | Stardance
|
anul_liyanage
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: WarioWare Game by @anul_liyanage | Stardanc Title: WarioWare Game by @anul_liyanage | Stardance...
|
1793735419.4822085
|
1794340219.4822085
|
|
67811
|
NULL
|
NULL
|
0
|
0.0
|
0
|
[]
|
[]
|
|
1793760643.116215
|
1794624643.116215
|
|
67810
|
Sticky by @KelvinY | Stardance
|
KelvinY
|
1
|
3.0
|
11
|
[{"ship_num": 1, "ts": "2026-1 [{"ship_num": 1, "ts": "2026-10-11T04:21:41Z", "hours": 3.0, "multiplier": 4.2, "payout": 11, "blessing": "neutral", "text": "Uploaded to itch.io!! you can now get your own sticky :) my biggest failure is that I forgot to disable the godot default start screen and this make it looks ugly 😭😭 More in README!", "repo": "https://github.com/kelvinyhf/sticky", "demo": "https://kelvinyhf.itch.io/sticky-pet"}]...
|
[{"post_id": 78631, "ts": "202 [{"post_id": 78631, "ts": "2026-10-11T04:18:11Z", "hours": 1.8703, "text": "My pet is done! Added a feature where when you double click it, it opens GitHub Gist for you to take note"}, {"post_id": 76605, "ts": "2026-10-04T04:49:45Z", "hours": 0.6958, "text": "Prototype Completed! took me 40 mins to complete the guide, but it’s interesting! The next step is to design my pet sprite. His name is Sticky, obviously he looks like a sticky note"}]...
|
Title: Sticky by @KelvinY | Stardance
Devlogs:
Pr Title: Sticky by @KelvinY | Stardance
Devlogs:
Prototype Completed! took me 40 mins to complete the guide, but it’s interesting! The next step is to design my pet sprite. His name is Sticky, obviously he looks like a sticky note
My pet is done! Added a feature where when you double click it, it opens GitHub Gist for you to take note
Ships:
Uploaded to itch.io!! you can now get your own sticky :) my biggest failure is that I forgot to disable the godot default start screen and this make it looks ugly 😭😭 More in README!...
|
1793839780.4587429
|
1793926180.4587429
|
|
67809
|
My own site! by @velvet_void | Stardance
|
velvet_void
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My own site! by @velvet_void | Stardance
|
1793735416.2360625
|
1794340216.2360625
|
|
67808
|
NULL
|
NULL
|
0
|
0.0
|
0
|
[]
|
[]
|
|
1793760641.3986783
|
1794624641.3986783
|
|
67807
|
NULL
|
NULL
|
0
|
0.0
|
0
|
[]
|
[]
|
|
1793760639.3591535
|
1794624639.3591535
|
|
67806
|
KumoWake by @Rendonut | Stardance
|
Rendonut
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: KumoWake by @Rendonut | Stardance
|
1793735414.1776874
|
1794340214.1776874
|
|
67805
|
Electric Vehicle by @zappy | Stardance
|
zappy
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Electric Vehicle by @zappy | Stardance
|
1793735412.0572424
|
1794340212.0572424
|
|
67804
|
NULL
|
NULL
|
0
|
0.0
|
0
|
[]
|
[]
|
|
1793760638.4899933
|
1794624638.4899933
|
|
67803
|
My awesome macropad! by @Lais | Stardance
|
Lais
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My awesome macropad! by @Lais | Stardance
|
1793735407.5696247
|
1794340207.5696247
|
|
67802
|
My own site! by @27_jamesclark | Stardance
|
27_jamesclark
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My own site! by @27_jamesclark | Stardance
|
1793735402.7608116
|
1794340202.7608116
|
|
67801
|
My awesome macropad! by @Aiknoor | Stardance
|
Aiknoor
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My awesome macropad! by @Aiknoor | Stardanc Title: My awesome macropad! by @Aiknoor | Stardance...
|
1793735396.2924423
|
1794340196.2924423
|
|
67800
|
BLARE by @waruingi | Stardance
|
waruingi
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: BLARE by @waruingi | Stardance
|
1793735390.5090506
|
1794340190.5090506
|
|
67799
|
truthscope by @bakshit624 | Stardance
|
bakshit624
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: truthscope by @bakshit624 | Stardance
|
1793735388.4736974
|
1794340188.4736974
|
|
67798
|
My awesome macropad! by @azureworksinc | Stardance
|
azureworksinc
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My awesome macropad! by @azureworksinc | St Title: My awesome macropad! by @azureworksinc | Stardance...
|
1793735386.3358834
|
1794340186.3358834
|
|
67797
|
Custom bobbleheads by @malcolmM | Stardance
|
malcolmM
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Custom bobbleheads by @malcolmM | Stardance
|
1793735385.444136
|
1794340185.444136
|
|
67796
|
Untitled project by @bakshit624 | Stardance
|
bakshit624
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Untitled project by @bakshit624 | Stardance
|
1793735384.5899477
|
1794340184.5899477
|
|
67795
|
Omni Directional Drivebase by @sircheetodust | Sta Omni Directional Drivebase by @sircheetodust | Stardance...
|
sircheetodust
|
0
|
2.36
|
0
|
[]
|
[{"post_id": 76852, "ts": "202 [{"post_id": 76852, "ts": "2026-10-04T20:18:08Z", "hours": 1.1431, "text": "Devlog 2! Welcome back, today I made the body of the robot and made some modifications to the adapters. I shortened the shaft adapters by 8 mm to stop the wheel from protruding out as much as it did. As for the base, I didnt do much extra just yet. Soon i am going to make 2 mounts for the L298N motor drivers which control the motor, a mount for the arduino, and comms SBC and buck converter."}, {"post_id": 76606, "ts": "2026-10-04T04:52:39Z", "hours": 1.2178, "text": "Devlog 01! I am planning to make an X drivetrain with omni wheels and 12 V 600 rpm motors. I started out by throwing stuff into onshape and hoping it works. So far I have this testing wheel assembly, which I will mount in an X shape like this diagram I definetly did not steal from the vex website. Electronics wise, I will have an arduino uno r4 + motor controller hat, and a rock 4 c+ for live coms via controller and telemetry to my laptop like FRC robots. Firmware wise, I have already have a coms system which relays laptop signals to the arduino and can allow for a over the air UI system. Arduino is also simple enough so I will spend the most time cadding this(no idea what im doing). wait onshape has dark mode?"}]...
|
Title: Omni Directional Drivebase by @sircheetodus Title: Omni Directional Drivebase by @sircheetodust | Stardance
Devlogs:
Devlog 01! I am planning to make an X drivetrain with omni wheels and 12 V 600 rpm motors. I started out by throwing stuff into onshape and hoping it works. So far I have this testing wheel assembly, which I will mount in an X shape like this diagram I definetly did not steal from the vex website. Electronics wise, I will have an arduino uno r4 + motor controller hat, and a rock 4 c+ for live coms via controller and telemetry to my laptop like FRC robots. Firmware wise, I have already have a coms system which relays laptop signals to the arduino and can allow for a over the air UI system. Arduino is also simple enough so I will spend the most time cadding this(no idea what im doing). wait onshape has dark mode?
Devlog 2! Welcome back, today I made the body of the robot and made some modifications to the adapters. I shortened the shaft adapters by 8 mm to stop the wheel from protruding out as much as it did. As for the base, I didnt do much extra just yet. Soon i am going to make 2 mounts for the L298N motor drivers which control the motor, a mount for the arduino, and comms SBC and buck converter....
|
1793735382.4912622
|
1794340182.4912622
|
|
67794
|
TeenyTinyStories by @yf | Stardance
|
yf
|
0
|
30.16
|
0
|
[]
|
[{"post_id": 78495, "ts": "202 [{"post_id": 78495, "ts": "2026-10-10T23:27:58Z", "hours": 4.4483, "text": "Evals are done Finally, after two days of tweaking values and rewriting evaluation prompts, the eval is all done. As per the original paper I employed a GPT-Eval adjacent score (albeit with Ministral-14B) as the LLM judge) and an Elo system (with GPT 5.6 Luna as the judge). The results did significantly corroborate with each other, which validates our conclusion that TTS-1M did as well or exceeded (slightly) as TS-1M whilst TTS-33M was poorer in performance compared to TS-33M though still competitive. Both the 33M models significantly outcompeted the 1M models, as expected and serves as an auxiliary validation."}, {"post_id": 77963, "ts": "2026-10-09T00:54:28Z", "hours": 4.7397, "text": "While we’re waiting…"}, {"post_id": 77744, "ts": "2026-10-08T04:38:53Z", "hours": 4.0003, "text": "Models are finished The models have officially been fully trained and are ready for evaluation. I trained a 1M and a 33M (nominal) parameter model on the TTS dataset; at validation, shows much promise. I also drafted the Methods section and the Related Works section. I will continue to work on the evaluation section and the discussion section, before writing the final paper."}, {"post_id": 77443, "ts": "2026-10-07T00:58:48Z", "hours": 4.3611, "text": "On the Models The 1M model is finished training with tts 1.9870 / 7.29 (4.02M tok), ts_cap 2.3691 / 10.69, ts_full 2.3852 / 10.86 . It cost me around 10 USD but I am happy with its results. The 33M model is now undergoing training with ETA in around 10 hours. The models themselves use a GPT-Neo architecture with the EleutherAI/GPT-Neo-125M tokenizer, as per the original paper. I also worked on the paper, namely the methodology section. I don’t know if I can do the 3/9/28M models however due to budget constraints but I will try."}, {"post_id": 76948, "ts": "2026-10-05T05:10:19Z", "hours": 3.3178, "text": "Devlog 3 Full corpus generation finished ASAP on 4090. You can find the TTS corpus on ableian/teenytinystories on Huggingface. Now, I followed the architecture specified by Eldan and Li and started training the 1M/3M and 9M models. Will finish in 16 hours, so I’m excited!"}, {"post_id": 76665, "ts": "2026-10-04T09:26:41Z", "hours": 3.1161, "text": "Devlog 2 In here, I introduce you to the official generation pipeline.I used the TinyStories-Instruct-33M checkpoint as the progenitor model. It accepts a structured header as follows: Words: verb, noun, adjective Features: X, Y [optional] Story: The words were chosen from the original TS word list from TinyStories-Instruct-valid.txt . The features were derived from an optional feature set, sampled from the empirical distributions in the same TinyStories-Instruct data. Quality control was also enforced by rejecting incomplete stories and stories that were too long or too short. I wrote an .ipynb and began generating stories in batches of 1024 on a rented 4090 (CUDA) and my M1 MacBook Pro (MPS)! Should be ready in around 8 hours."}, {"post_id": 76587, "ts": "2026-10-04T02:34:54Z", "hours": 6.1778, "text": "1 On TinyStories TinyStories (Eldan & Li, 2023) represented a significant breakthrough in the field of language model training, providing evidence that an synthetic (i.e. LLM-generated) corpus could be used to efficiently train a Small Language Model (SLM) fluent in English and basic story generation. The pipeline is direct: Eldan and Li seeded ChatGPT (3.5 and 4) with story generation prompts, generating over 470 M tokens (~2.1 million stories), before training a 1, 3, 9, 28 and 33M SLM. Through a range of evaluation metrics, they demonstrated that these models could output fluent English, and for larger and more layered models, could generate semi-coherent stories. 2 On TeenyTinyStories Now, I propose a step further: what happens if we take the Instruct variant of the official checkpoint, TinyStories-Instruct-33M instead of a frontier model like the GPT series, to generate a new corpus of data and train the same models on this corpus instead? This is the core of this project.Note: I understand that this is fundamentally a research project rather than produce-a-usable-software type thing you see on Stardance, so I will produce a bunch of usables, including a conversation-tuned instruct variant of the 33M which you can chat with in your browser in the form of a website. The core results, however, will be in the form of an ArXiV preprint before I submit to NHSJS. What I did After I skimmed through the 2023 paper, I summarised their corpus generation pipeline as follows:The generator used GPT-3.5 and GPT-4 via the OpenAI APIPrompt composition composed of a list of required words (one verb, one noun, one adjective), and an (optional) story feature (i.e. dialogue, bad ending, moral value, plot twist…) to generate a story. Example: “Write a short story (3-5 paragraphs) which only uses very simple words that a 3 year old child would likely understand.The story should use the verb “decorate”,the noun “thunder”and the adjective “ancient”.The story should have the following features:the story should contain at least one dialogue,the story has a bad ending.Remember to only use simple words!”Accordingly, I wrote the draft introduction and the Related Works section, and planned the Methodology."}]...
|
Title: TeenyTinyStories by @yf | Stardance
Devlog Title: TeenyTinyStories by @yf | Stardance
Devlogs:
1 On TinyStories TinyStories (Eldan & Li, 2023) represented a significant breakthrough in the field of language model training, providing evidence that an synthetic (i.e. LLM-generated) corpus could be used to efficiently train a Small Language Model (SLM) fluent in English and basic story generation. The pipeline is direct: Eldan and Li seeded ChatGPT (3.5 and 4) with story generation prompts, generating over 470 M tokens (~2.1 million stories), before training a 1, 3, 9, 28 and 33M SLM. Through a range of evaluation metrics, they demonstrated that these models could output fluent English, and for larger and more layered models, could generate semi-coherent stories. 2 On TeenyTinyStories Now, I propose a step further: what happens if we take the Instruct variant of the official checkpoint, TinyStories-Instruct-33M instead of a frontier model like the GPT series, to generate a new corpus of data and train the same models on this corpus instead? This is the core of this project.Note: I understand that this is fundamentally a research project rather than produce-a-usable-software type thing you see on Stardance, so I will produce a bunch of usables, including a conversation-tuned instruct variant of the 33M which you can chat with in your browser in the form of a website. The core results, however, will be in the form of an ArXiV preprint before I submit to NHSJS. What I did After I skimmed through the 2023 paper, I summarised their corpus generation pipeline as follows:The generator used GPT-3.5 and GPT-4 via the OpenAI APIPrompt composition composed of a list of required words (one verb, one noun, one adjective), and an (optional) story feature (i.e. dialogue, bad ending, moral value, plot twist…) to generate a story. Example: “Write a short story (3-5 paragraphs) which only uses very simple words that a 3 year old child would likely understand.The story should use the verb “decorate”,the noun “thunder”and the adjective “ancient”.The story should have the following features:the story should contain at least one dialogue,the story has a bad ending.Remember to only use simple words!”Accordingly, I wrote the draft introduction and the Related Works section, and planned the Methodology.
Devlog 2 In here, I introduce you to the official generation pipeline.I used the TinyStories-Instruct-33M checkpoint as the progenitor model. It accepts a structured header as follows: Words: verb, noun, adjective Features: X, Y [optional] Story: The words were chosen from the original TS word list from TinyStories-Instruct-valid.txt . The features were derived from an optional feature set, sampled from the empirical distributions in the same TinyStories-Instruct data. Quality control was also enforced by rejecting incomplete stories and stories that were too long or too short. I wrote an .ipynb and began generating stories in batches of 1024 on a rented 4090 (CUDA) and my M1 MacBook Pro (MPS)! Should be ready in around 8 hours.
Devlog 3 Full corpus generation finished ASAP on 4090. You can find the TTS corpus on ableian/teenytinystories on Huggingface. Now, I followed the architecture specified by Eldan and Li and started training the 1M/3M and 9M models. Will finish in 16 hours, so I’m excited!
On the Models The 1M model is finished training with tts 1.9870 / 7.29 (4.02M tok), ts_cap 2.3691 / 10.69, ts_full 2.3852 / 10.86 . It cost me around 10 USD but I am happy with its results. The 33M model is now undergoing training with ETA in around 10 hours. The models themselves use a GPT-Neo architecture with the EleutherAI/GPT-Neo-125M tokenizer, as per the original paper. I also worked on the paper, namely the methodology section. I don’t know if I can do the 3/9/28M models however due to budget constraints but I will try.
Models are finished The models have officially been fully trained and are ready for evaluation. I trained a 1M and a 33M (nominal) parameter model on the TTS dataset; at validation, shows much promise. I also drafted the Methods section and the Related Works section. I will continue to work on the evaluation section and the discussion section, before writing the final paper.
While we’re waiting…
Evals are done Finally, after two days of tweaking values and rewriting evaluation prompts, the eval is all done. As per the original paper I employed a GPT-Eval adjacent score (albeit with Ministral-14B) as the LLM judge) and an Elo system (with GPT 5.6 Luna as the judge). The results did significantly corroborate with each other, which validates our conclusion that TTS-1M did as well or exceeded (slightly) as TS-1M whilst TTS-33M was poorer in performance compared to TS-33M though still competitive. Both the 33M models significantly outcompeted the 1M models, as expected and serves as an auxiliary validation....
|
1793735380.2204483
|
1794340180.2204483
|