Osakra/Project-Norn-V18-9B
Text Generation • 9B • Updated • 83 • 6
sometimesanotion
sometimesanotion
AI & ML interests
Agentic LLM services, model merging, finetunes, distillation
Recent Activity
liked a model about 11 hours ago
Osakra/Project-Norn-V18-9B liked a model about 23 hours ago
Q1z/Pivot liked a model 2 days ago
bespokelabs/Bespoke-Nimble-9BOrganizations
liked a
model about 11 hours ago
liked a
model about 23 hours ago
liked a
model 2 days ago
liked 2
models 3 days ago
New activity in DavidAU/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-ULTRA-HERETIC-Uncensored-NM-DAU-NEO-MTP-GGUF 4 days ago
A few generation issues with creative writing...
12
#4 opened 5 days ago
by 
BTGbullseye
liked a
model 4 days ago
reacted to SeaWolf-AI's post with 🧠🔥 6 days ago
Post
5631
Instead of making the fly brain play games, we measured what it is for
Since the Drosophila connectome was released, people have had the fly brain doomscroll a feed, play Beat Saber, drive in GTA. Those demos show that the brain runs. We wanted to show what it is for.
So we gave it a looming object — one of the few things a fly brain is unambiguously built to detect — then deleted a single cell type and repeated the identical stimulus. Remove LC4, 126 cells out of 173,023, and the escape signal falls from 0.840 to 0.091. Eighty-nine percent of the danger signal is gone while the other 172,897 neurons run exactly as before.
Deleting neurons does not do this on its own, which is the whole point of the controls. LC11 is the same class and larger than LC4 — 143 cells and 9,940 outgoing connections against 126 and 7,846 — and removing every one of them changes the signal by 0.000000, to six decimal places. It has to be those 126.
No server and no GPU: a looming stimulus drives fewer than one percent of neurons above threshold, so the whole thing is 40 KB gzipped and runs in your browser.
Try deleting one yourself →
VIDraft/fruitfly-brain
The wiring is the measured connectome, but synaptic strength is a uniform count-based value and the dynamics are a firing-rate model of our choosing — a total-effect measurement of a model, not a recording from a fly. Male CNS connectome, FlyEM / HHMI Janelia with Google Research, Columbia and Harvard (2026), CC BY.
Since the Drosophila connectome was released, people have had the fly brain doomscroll a feed, play Beat Saber, drive in GTA. Those demos show that the brain runs. We wanted to show what it is for.
So we gave it a looming object — one of the few things a fly brain is unambiguously built to detect — then deleted a single cell type and repeated the identical stimulus. Remove LC4, 126 cells out of 173,023, and the escape signal falls from 0.840 to 0.091. Eighty-nine percent of the danger signal is gone while the other 172,897 neurons run exactly as before.
Deleting neurons does not do this on its own, which is the whole point of the controls. LC11 is the same class and larger than LC4 — 143 cells and 9,940 outgoing connections against 126 and 7,846 — and removing every one of them changes the signal by 0.000000, to six decimal places. It has to be those 126.
No server and no GPU: a looming stimulus drives fewer than one percent of neurons above threshold, so the whole thing is 40 KB gzipped and runs in your browser.
Try deleting one yourself →
VIDraft/fruitfly-brain
The wiring is the measured connectome, but synaptic strength is a uniform count-based value and the dynamics are a firing-rate model of our choosing — a total-effect measurement of a model, not a recording from a fly. Male CNS connectome, FlyEM / HHMI Janelia with Google Research, Columbia and Harvard (2026), CC BY.
liked a
Space 6 days ago
liked 2
models 9 days ago
Best performing model of qwen3.6 family
🚀🤗 1
39
#3 opened 25 days ago
by 
repla73
liked a
model 9 days ago
Can this technique be applied to Qwen 3.8 Flash Next?
6
#30 opened 19 days ago
by 
pohnean
liked a
model 14 days ago
Merges with Qwen3.8-27B?
18
#3 opened about 1 month ago
by sometimesanotion
liked a
model 18 days ago
New activity in DavidAU/Qwen3.6-40B-Claude-4.6-Opus-Deckard-Heretic-Uncensored-Thinking-NEO-CODE-Di-IMatrix-MAX-GGUF 23 days ago
Maybe expand LiquidAI models?
❤️ 1
1
#33 opened 23 days ago
by sometimesanotion
I was hoping for a prune/distill of this model!
2
#1 opened 25 days ago
by sometimesanotion