INDEX

Explanations

Code and technical language

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

 rock

-0.30

 compartment

-0.27

 stone

-0.27

 Rock

-0.26

olith

-0.26

edb

-0.26

ed

-0.25

ulus

-0.25

 Stone

-0.25

ä¸ĩåĲ¨

-0.25

POSITIVE LOGITS

æī°ä¹±

0.27

 smoothed

0.26

ADF

0.26

 polished

0.25

 mouths

0.25

smooth

0.24

è¿ĲèĲ¥ç®¡çĲĨ

0.24

æµĭ

0.24

æµģå¤±

0.24

hf

0.23

Activations Density 0.001%

Code and technical language

No Comments

No Known Activations