INDEX

Explanations

symmetric/asymmetric

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_11.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.11.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

åĨł

-0.29

æĿ¿

-0.28

æľ¬ç¶²ç«Ļ

-0.26

))*(

-0.26

_STRUCTURE

-0.26

ç¡¼

-0.25

 boards

-0.25

å»Ĭ

-0.25

 Corps

-0.24

é¡ĺ

-0.24

POSITIVE LOGITS

gain

0.29

æ¾ľ

0.27

StateChanged

0.26

ç»Ĳ

0.26

Gain

0.25

ä¸ªå¤ļ

0.25

åŁı

0.25

æ¯Ķè¾ĥå¤ļ

0.24

//{↵

0.24

 gains

0.24

Activations Density 0.019%

symmetric/asymmetric

No Comments

No Known Activations