Skip to content

arianraje

qwen3-4b-mamba3-hybrid-stage1-align-bias0

arianraje/qwen3-4b-mamba3-hybrid-stage1-align-bias0

qwen3-4b-mamba3-hybrid-stage1-align-bias0 is a text generation model from arianraje. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as apache-2.0.

Stage-1 endpoint, bias-0 arm -- per-layer hidden-state alignment, 100M tokens at sequence length 4096, mixers only (LR 1e-3, micro-batch 16, 1,526 steps on 4x B200). Each converted layer takes the teacher's hs[i] and…

Downloads · 30 days

13

3% of all-time downloads

All-time downloads

398

Public

Parameters

4.7B

9.5 GB on disk

Likes

0

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors9.5 GB · 100%

At a glance

Task
Text Generation
Library
transformers
License
apache-2.0
Model type
qwen3_mamba3
Access
Public
Created
Sep 10, 2026
Updated
Sep 11, 2026
SHA
8958d5cf

Try a prompt

Base models

Task
Text Generation
Library
transformers
Type
qwen3_mamba3
License
apache-2.0
Created
Sep 10, 2026
Updated
Sep 11, 2026