INDEX

Explanations

neut

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

wz

-0.31

 flag

-0.25

æĮº

-0.25

are

-0.25

 flagged

-0.25

ÐºÐ»Ð°Ð´

-0.24

ØŃÙĤ

-0.24

è®¸

-0.24

 deflate

-0.24

æłİ

-0.23

POSITIVE LOGITS

yg

0.25

ãĥĥãĥĦ

0.25

tics

0.24

 Spice

0.24

elight

0.23

trail

0.23

ramer

0.23

ign

0.23

æľºèº«

0.23

éķĺ

0.23

Activations Density 0.001%

neut

No Known Activations