INDEX

Explanations

Nonsense

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_23.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.23.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 dash

-0.26

åĩºåħ¥

-0.26

KHTML

-0.25

ä¾Ŀæ³ķè¿½ç©¶

-0.24

<main

-0.24

èĥĮä¸Ĭ

-0.24

.prof

-0.23

ffi

-0.23

umph

-0.23

.gwt

-0.23

POSITIVE LOGITS

iger

0.29

ier

0.29

_SIG

0.28

SIG

0.25

Rig

0.25

à¹Ģà¸¥

0.25

sig

0.25

lection

0.24

 Classic

0.24

ç¼ĸçłģ

0.24

Activations Density 0.001%

Nonsense

No Comments

No Known Activations