asigalov61/Music-Llama-4
Updated • 2
Alex
asigalov61
AI & ML interests
Music AI
Recent Activity
updated a model 8 days ago
asigalov61/Music-Llama-4 new activity 8 days ago
asigalov61/Music-Llama-4:Missing model card and configuration updated a dataset 8 days ago
asigalov61/misc-and-tempOrganizations
updated a
model 8 days ago
Missing model card and configuration
❤️ 1
1
#1 opened 8 days ago
by 
korigamik
updated a
dataset 8 days ago
Create base_large_fine_tuned/README.md
#1 opened 11 days ago
by 
projectlosangeles
updated a
model 11 days ago
liked a
Space 11 days ago
reacted to Avifenesh's post with 🚀❤️ 11 days ago
Post
2,180
I added a small gate to the Gated DeltaNet layers of Qwen3.5 and trained it with GRPO under a length penalty.
On Qwen3.5-4B it reasons 48% shorter than LoRA on MATH-500, and anytime accuracy goes from 0.786 to 0.852. A LoRA with about the same parameter count barely moves.
The gate sits on beta, the per-token write strength. One matrix per layer, zero init, so it starts as a no-op.
It started as a negative eigenvalue experiment. The negative range turned out not to matter. The gate did.
Write-up: https://huggingface.co/blog/Avifenesh/beta-gate-qwen35
Adapters: Avifenesh/gate-qwen35-6accb63081800b2f2d583b80
Code: https://github.com/avifenesh/gdn-beta-gatehttps://huggingface.co/collections/Avifenesh/gate-qwen35
On Qwen3.5-4B it reasons 48% shorter than LoRA on MATH-500, and anytime accuracy goes from 0.786 to 0.852. A LoRA with about the same parameter count barely moves.
The gate sits on beta, the per-token write strength. One matrix per layer, zero init, so it starts as a no-op.
It started as a negative eigenvalue experiment. The negative range turned out not to matter. The gate did.
Write-up: https://huggingface.co/blog/Avifenesh/beta-gate-qwen35
Adapters: Avifenesh/gate-qwen35-6accb63081800b2f2d583b80
Code: https://github.com/avifenesh/gdn-beta-gatehttps://huggingface.co/collections/Avifenesh/gate-qwen35
liked a
Space about 1 month ago
updated a
Space about 1 month ago
liked 2
Spaces about 1 month ago
liked 5
Spaces 2 months ago
DDSP-Guitar
🎸
2
Synthesize guitar audio from MIDI with DDSP
Stable Audio 3
✏
2
Restyle, inpaint, or continue music from a prompt
CrisperWhisper 2.0 Turbo
🎙
2
Verbatim & intended speech recognition with word timestamps
YingMusic-Singer
🎤
22
Zero-shot singing voice synthesis & lyric editing
Tafsut Forecast
📈
7
Zero-shot probabilistic time-series forecasting
updated a
dataset 2 months ago
liked a
model 2 months ago
liked a
Space 2 months ago