INDEX

Explanations

location, placement

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_20.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.20.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

akin

-0.30

ç¬Ķ

-0.27

pen

-0.25

ãĢħ

-0.25

 Tues

-0.25

pen

-0.24

 rate

-0.24

="__

-0.24

ppers

-0.24

stead

-0.24

POSITIVE LOGITS

åĢŁ

0.28

 limp

0.27

 broadcasters

0.26

.lin

0.26

æĮĩçĶ²

0.26

ä¹ĸ

0.25

ä¸»æĦı

0.25

å¸Ľ

0.25

.rf

0.24

åīģ

0.24

Activations Density 10.767%

location, placement

No Known Activations