INDEX

Explanations

of

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_25.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.25.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

åĳ

-0.28

å£«

-0.26

åĳĺ

-0.26

cursor

-0.26

 domestically

-0.26

 Cursor

-0.25

æīĭèĦļ

-0.25

 scalar

-0.25

syscall

-0.25

 autonom

-0.25

POSITIVE LOGITS

ç©´

0.30

ä¸ºå¯¼åĲĳ

0.27

Unused

0.27

éĺĶ

0.26

 State

0.25

 Expanded

0.25

isis

0.25

æŃ¤æĸĩ

0.24

verse

0.24

ç§¯èĵĦ

0.24

Activations Density 0.114%

of

No Comments

No Known Activations