INDEX

Explanations

Harlem

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

å°¾

-0.29

 tail

-0.29

tail

-0.28

Ð¿ÑĥÑģ

-0.25

è¯Ŀè¯Ń

-0.25

ç»ĦåĽ¾

-0.25

 sparing

-0.24

ë²Į

-0.24

licant

-0.24

meric

-0.24

POSITIVE LOGITS

å½±

0.27

tran

0.26

èĮĦåŃĲ

0.26

rita

0.26

èı¡

0.25

ahi

0.25

Boy

0.25

 Town

0.25

åĨĴ

0.24

 Shadows

0.24

Activations Density 0.001%

Harlem

No Comments

No Known Activations