INDEX

Explanations

ghost

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

iani

-0.31

èĢ³æľµ

-0.31

igte

-0.30

 more

-0.28

rouw

-0.27

éĺģ

-0.27

æĽ´å¤ļ

-0.26

 MORE

-0.26

asha

-0.25

 mehr

-0.25

POSITIVE LOGITS

ing

0.28

éĵĥ

0.26

å¹³çŃī

0.25

Son

0.25

zell

0.25

åĲĮä¸ļ

0.24

 interaction

0.24

oval

0.24

getFullYear

0.24

è¯ģæĺİ

0.24

Activations Density 0.032%

ghost

No Comments

No Known Activations