Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Raise a neural network as a pet - a transparent cognitive sandbox where a digital squid learns, grows, and rewires itself via Hebbian learning & Neurogenesis
| Date | Stars |
|---|---|
| 2026-08-05 | 304 |
| 2026-08-06 | 304 |
| 2026-08-10 | 305 |
| 2026-08-11 | 306 |
| 2026-08-14 | 306 |
| 2026-08-15 | 306 |
| 2026-08-17 | 307 |
| 2026-08-18 | 306 |
| 2026-08-20 | 307 |
| 2026-08-23 | 308 |
| 2026-08-25 | 310 |
| 2026-08-26 | 311 |
| 2026-08-28 | 312 |
| 2026-08-29 | 313 |
| 2026-09-01 | 313 |
| 2026-09-12 | 313 |
| 2026-09-15 | 314 |
| 2026-09-26 | 317 |
| 2026-09-29 | 318 |
| 2026-10-01 | 319 |
| 2026-10-02 | 320 |
| 2026-10-05 | 321 |
| 2026-10-07 | 321 |
| 2026-10-10 | 322 |
| 2026-10-11 | 323 |
| 2026-10-12 | 324 |
| 2026-10-17 | 324 |
| 2026-10-18 | 325 |
| 2026-10-20 | 326 |
| 2026-10-24 | 327 |
| 2026-10-26 | 329 |
| 2026-11-01 | 330 |
| 2026-11-04 | 331 |
Today
+1 stars today
This week
+2 stars this week
This month
+10 stars this month
Momentum
21.0
growth rate 0.61%/day
_"What if a Tamagotchi had a neural network and could learn stuff?"_ - [Gigazine](https://gigazine.net/gsc_news/en/20250505-dosidicus-electronicae/) , [Hackaday](https://hackaday.com/2025/04/26/digital-squids-behavior-shaped-by-neural-network/)
<p align="left">
<img src="https://img.shields.io/badge/AI-Neural_Network-9C27B0?style=flat&logo=mindmeister&logoColor=white" height="20" alt="AI">
<img src="https://img.shields.io/badge/License-GPL_v2-blue.svg?style=flat" height="20" alt="GPL-2.0">
<img src="https://img.shields.io/badge/Translations-7-228B22?style=flat&logo=google-translate&logoColor=white&labelColor=333333" height="20" alt="Translations">
<a href="https://buymeacoffee.com/vicioussquid"><img src="https://img.shields.io/badge/Buy_Me_A_Coffee-FFDD00?style=flat&logo=buy-me-a-coffee&logoColor=black" height="20" alt="Buy Me A Coffee"></a>
</p>
> ### #1 `weird` topic project on GitHub — and #2 `simulation-game`.
# _Dosidicus electronicus_
### Learn neuroscience by **raising a neural network as a pet**
_A transparent cognitive sandbox disguised as a digital pet squid with a neural network you can **see thinking**_
- Part **educational neuro tool**, part **sim game**, part **fever dream**
- Combining 1990s virtual pet nostalgia with modern computational neuroscience.
### [Download for Windows, Linux and Android](https://github.com/ViciousSquid/Dosidicus/releases)
```bash
git clone https://github.com/ViciousSquid/Dosidicus.git
cd Dosidicus
python -m venv venv
source venv/bin/activate # or venv\Scripts\activate (Windows)
pip install -r requirements.txt
python main.py
```
<img src="https://github.com/user-attachments/assets/02119926-47f7-4bfb-96b9-457d470064e4" width="900">
<img src="https://github.com/user-attachments/assets/496cec0d-0810-4f47-8618-11165e0dd50d" width="380">
---
## [Manifesto](https://github.com/ViciousSquid/Dosidicus/wiki/Cognitive-Sandbox-Manifesto-%7C-Artificial-Life-and-Transparent-Neural-Systems) | [Changelog](https://github.com/ViciousSquid/Dosidicus/wiki/changelog) | [Wiki](https://github.com/ViciousSquid/Dosidicus/wiki) (57 pages)
---
## **Why this exists**
What if you could understand every neuron inside a learning creature?
The project is designed to make artificial cognition visible.
Instead of hiding intelligence inside millions of parameters, Dosidicus starts with just eight core neurons: seven drives and the one sense a squid cannot live without. The brain is drawn in labelled rows — the drives across the top, the motor bank underneath them (one neuron per thing the squid can do, driven by ordinary synapses rather than by a rule), and the senses below that. Every connection can be inspected. Every activation can be visualised. Every learned behaviour can be traced back to experience.
As the squid lives, its brain rewires itself through Hebbian learning, strengthens useful pathways using [STDP](https://github.com/ViciousSquid/Dosidicus/wiki/Spike%E2%80%90Timing%E2%80%90Dependent-Plasticity-(STDP)), works out which of its own actions cause which consequences, and grows entirely new neurons when its existing structure cannot cope.
And it can tell you all of it. Every synaptic change and every grown neuron records **why** — the mechanism, the evidence, the experience, the action, the consequence — and the brain tools read that record rather than reconstructing an approximation of it:
> *"Why did this weight change from 0.31 to 0.47?"*
> *"Why does this neuron exist?"*
> *"What does the squid know about food?"*
No two brains ever develop the same way.
## "Why believe it?"
**Not faked:** there is [one learning engine](https://github.com/ViciousSquid/Dosidicus/blob/3.1_LatestVersion/src/plasticity.py) shared by the GUI, the headless trainer, and the test suite — the squid cannot learn differently depending on which interface watches it. Every weight change records its mechanism, evidence, and triggering experience.
## As the caretaker you will
- Feed, clean and care for Excerpt of 10,742 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:f7776a2d524bbb8c, llm:Repository description and topics: 'Raise a neural network as a pet - ... Hebbian learning & Neurogenesis', topics include neural-networks, hebbian-learning, neurogenesis, neuroevolution, neuron-simulator, neuroscience, virtual-pet, simulation-game; README describes educational neuro tool and simulation game.
matched fp:f7776a2d524bbb8c, llm:Repository description and topics: 'Raise a neural network as a pet - ... Hebbian learning & Neurogenesis', topics include neural-networks, hebbian-learning, neurogenesis, neuroevolution, neuron-simulator, neuroscience, virtual-pet, simulation-game; README describes educational neuro tool and simulation game.