INDEX

Explanations

crypt

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

bet

-0.30

porte

-0.28

sthrough

-0.28

èį·

-0.26

Bet

-0.26

ctica

-0.26

 Alto

-0.25

AMES

-0.25

sson

-0.24

 forth

-0.24

POSITIVE LOGITS

soc

0.30

æĢ§å¼º

0.27

erc

0.27

é«ĺæĺĤ

0.27

'),'

0.26

cow

0.26

-)

0.25

fax

0.25

éļľ

0.25

svc

0.24

Activations Density 0.248%

crypt

No Comments

No Known Activations