Downloads · 30 days
32
12% of all-time downloads
Neo-Qwen3-4B-Instruct-2507-v0
neopolita/Neo-Qwen3-4B-Instruct-2507-v0
Neo-Qwen3-4B-Instruct-2507-v0 is a machine learning model from neopolita. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.
<img alt="image" src="https://huggingface.co/neopolita/qwen3-4b-instruct-2507-neopolitized-v0/resolve/main/alchemy.png" width="500"
All-time downloads
262
Public
Repo size
16.1 GB
Likes
1
Public
Repo makeup
Click a slice to open those files.
.gguf8.1 GB · 100%
At a glance
- License
- apache-2.0
- Access
- Public
- Created
- Nov 17, 2025
- Updated
- Dec 28, 2025
- SHA
- d1decdb0
Base models
Model card
From the Hugging Face model README
<img alt="image" src="https://huggingface.co/neopolita/qwen3-4b-instruct-2507-neopolitized-v0/resolve/main/alchemy.png" width="500">
<font color="cccccc">qwen3-4b-instruct-2507 neopolitized with projected shards and fragments of openai/gpt-oss-20b.</font>
- projection method: 1, {c: 0}, no_c, no_n
- merge method: 1
- layers: 0-14 [x->x]
- alpha: 0.85-0.85
- tensors: attn_k.T, attn_v.T, attn_q.T | Metric | Delta | |------------|------:| | MMLU 400 | +0.50 | | Winogrande | +0.55 | | Hellaswag | +0.75 |
8 w w
8d8b. .d88b .d8b. 88b. .d8b. 8 w w8ww .d88
8P Y8 8.dP' 8' .8 8 8 8' .8 8 8 8 8 8
8 8 `Y88P `Y8P' 88P' `Y8P' 8 8 Y8P `Y88
8
- License
- apache-2.0
- Created
- Nov 17, 2025
- Updated
- Dec 28, 2025