INDEX

Explanations

sand

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_2.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.2.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

unp

-0.28

 organisms

-0.27

eur

-0.26

UNU

-0.25

 grey

-0.24

},{

-0.24

 crafts

-0.24

ecast

-0.24

spl

-0.24

 uncert

-0.23

POSITIVE LOGITS

onChange

0.27

hang

0.27

 possible

0.27

retval

0.26

 agon

0.25

optional

0.25

åĲĥä»Ģä¹Ī

0.25

aris

0.25

really

0.25

ç»ĨåĪĻ

0.25

Activations Density 32.312%

sand

No Comments

No Known Activations