INDEX

Explanations

Scientific and news articles

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

icum

-0.30

thur

-0.30

å¹ķåĲİ

-0.28

Apt

-0.27

 behind

-0.26

/apt

-0.26

iap

-0.25

rib

-0.25

.struts

-0.25

æŀ¢

-0.25

POSITIVE LOGITS

çĶ¾

0.26

pipes

0.26

_NATIVE

0.25

lul

0.25

Lemma

0.24

+N

0.24

ext

0.24

ERGE

0.23

ugo

0.23

+W

0.23

Activations Density 0.006%

Scientific and news articles

No Comments

No Known Activations