INDEX

Explanations

with

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_7.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.7.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

.epsilon

-0.25

uet

-0.25

adier

-0.24

Ĺı

-0.24

ussen

-0.24

svp

-0.24

ervo

-0.24

IENTATION

-0.24

-action

-0.24

 action

-0.23

POSITIVE LOGITS

ha

0.27

 dÃ¼rfen

0.26

èĪ·

0.26

rim

0.25

 darf

0.24

æĭĻ

0.24

Anc

0.24

 secular

0.24

ä»Ģä¹Īäºĭ

0.24

ç´łæĿĲ

0.23

Activations Density 0.005%

with

No Comments

No Known Activations