INDEX

Explanations

Non-English words

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_27.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.27.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

æīĢæľīçļĦ

-0.38

çļĦæīĢæľī

-0.38

 about

-0.34

æīĢæľī

-0.31

 toutes

-0.31

åħ³äºİ

-0.31

çļĦä¸ĢåĪĩ

-0.30

about

-0.29

 every

-0.29

éĹľæĸ¼

-0.29

POSITIVE LOGITS

ç¼ºå°ĳ

0.30

è¿ĻçĤ¹

0.28

bsp

0.27

 haste

0.27

äººåĳĺ

0.26

sez

0.26

 ÄĳÃ¢u

0.26

 falta

0.26

è¿Ļä¸ĢçĤ¹

0.25

ä¸įåĪ°ä½į

0.25

Activations Density 0.002%

Non-English words

No Comments

No Known Activations