INDEX

Explanations

do

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

Negative Logits

ÑĥÐ¹

-0.31

 survey

-0.27

hem

-0.25

ä¸ºä¸»ä½ĵ

-0.24

legg

-0.24

éĺ¶æ®µæĢ§

-0.24

etus

-0.24

è¿ĻçĤ¹

-0.23

 book

-0.23

book

-0.23

POSITIVE LOGITS

RuleContext

0.34

anness

0.31

 StObject

0.28

 furnished

0.25

åı¥

0.25

frau

0.24

IBOutlet

0.24

imeType

0.24

åħ¥

0.23

FML

0.23

Activations Density 0.011%

do

No Known Activations