INDEX

Explanations

Code or data

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 conservatism

-0.30

 Quarter

-0.27

èģĶåĲĪä¼ļ

-0.27

 qualitative

-0.26

æĹłæīĢ

-0.26

ķìĿ¸

-0.25

è¿ĺåı¯ä»¥

-0.25

erie

-0.25

 conserv

-0.25

ierz

-0.24

POSITIVE LOGITS

åħļé£İ

0.30

 eigentlich

0.26

åıĳè¾¾

0.25

åĪĨéĴŁåĲİ

0.25

éģĵè·¯ä¸Ĭ

0.25

å¼ķæĿ¥

0.24

äº§ä¸ļç»ĵæŀĦ

0.24

 Emin

0.24

çİ¯å¢ĥæ±¡æŁĵ

0.24

Adjacent

0.24

Activations Density 0.006%

Code or data

No Comments

No Known Activations