CoolFace
Datasetpublic

juiceb0xc0de/Qwen3.5-2B-Base

juiceb0xc0de/Qwen3.5-2B-Base A brain atlas for Qwen/Qwen3.5-2B-Base, a 24-layer hybrid that runs linear attention on 18 layers and full attention on the other 6. This is not a chat dataset or a benchmark. It is an internal-mechanics map built by running activations through a corpus of prompts and scoring what each layer, component, head, and feature direction is doing. This is a base model, before any instruction tuning, so whatever structure shows up here was put there by… See the full description on the dataset page: https://huggingface.co/datasets/juiceb0xc0de/Qwen3.5-2B-Base.

sourceHugging Faceapache-2.0updated 11d agoView on Hugging Face
0likes537downloads
5 commits on main
f64658d11d ago

Declare Parquet splits so the viewer indexes; add banner

juiceb0xc0de
36a44c211d ago

Add Parquet tables for Dataset Viewer / SQL Console

juiceb0xc0de
9541e3f2mo ago

Upload README.md with huggingface_hub

juiceb0xc0de
01924013mo ago

Upload folder using huggingface_hub

juiceb0xc0de
30e5c9a3mo ago

initial commit

juiceb0xc0de