INDEX

Explanations

Diverse geographical locations/events

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

obel

-0.31

æĲŃéħį

-0.27

å¼Ĭ

-0.25

onation

-0.24

 emulation

-0.24

Cases

-0.23

éĿ¥

-0.23

 comprom

-0.23

åĤ¬

-0.23

ë²¨

-0.23

POSITIVE LOGITS

ç¼©

0.30

çº¸ä¸Ĭ

0.28

 road

0.27

road

0.26

çľŁç©º

0.26

çº¿ç´¢

0.26

è·¯

0.26

aren

0.25

ä¿Ŀå®Ī

0.25

.strip

0.24

Activations Density 0.013%

Diverse geographical locations/events

No Comments

No Known Activations