INDEX

Explanations

is

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

 about

-0.28

umin

-0.26

About

-0.26

rape

-0.25

æĮĸæİĺ

-0.25

egov

-0.25

/XML

-0.24

radient

-0.24

abeled

-0.24

 About

-0.24

POSITIVE LOGITS

Suite

0.28

 Suit

0.27

states

0.27

va

0.26

 suit

0.26

suite

0.26

_suite

0.25

Ð²Ð¸Ñģ

0.25

ä¸įå°½

0.24

va

0.24

Activations Density 0.007%

is

No Comments

No Known Activations