INDEX

Explanations

keratin

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_11.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.11.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

uper

-0.29

mise

-0.27

tor

-0.27

 tÃ©lÃ©

-0.27

 teased

-0.26

pur

-0.26

tractive

-0.25

CEL

-0.25

missive

-0.24

 update

-0.24

POSITIVE LOGITS

assi

0.30

 kÃ¶n

0.28

nst

0.25

ographies

0.25

 resp

0.25

olk

0.25

iid

0.25

èĽĨ

0.25

rots

0.24

CAC

0.24

Activations Density 0.067%

keratin

No Comments

No Known Activations