INDEX

Explanations

numbers and code

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

afari

-0.27

çŀ¿

-0.25

 ward

-0.25

@endforeach

-0.25

 pressures

-0.24

æ½ľèĥ½

-0.24

-pressure

-0.24

è³ĩ

-0.24

 pressure

-0.24

odom

-0.24

POSITIVE LOGITS

åıįæĦŁ

0.31

ä¸įå¼Ģ

0.29

åĺ´

0.27

éª¨å¤´

0.25

æĥ³äºĨ

0.25

gies

0.25

 sÃłng

0.25

otics

0.24

 demonstr

0.24

ritic

0.24

Activations Density 0.001%

numbers and code

No Comments

No Known Activations