INDEX

Explanations

virtue

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_22.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.22.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

GrÃ¼n

-0.24

æŁŀ

-0.23

feito

-0.23

gerÃ¤t

-0.23

éĽİ

-0.23

 Ð²Ð¾ÐºÑĢÑĥÐ³

-0.23

lar

-0.22

å®¤å¤ĸ

-0.22

ä½İåİĭ

-0.22

Elk

-0.22

POSITIVE LOGITS

hub

0.30

fulness

0.29

ç§ī

0.28

TY

0.28

blas

0.26

dsl

0.26

bl

0.26

less

0.25

dll

0.25

Du

0.24

Activations Density 0.033%

virtue

No Comments

No Known Activations