INDEX

Explanations

news and academic articles

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_19.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.19.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

jo

-0.29

ç¦ģ

-0.27

 patri

-0.27

 íĮ

-0.26

 Pear

-0.25

æµģè½¬

-0.25

ulo

-0.25

vie

-0.24

å¿Į

-0.24

 è¡

-0.24

POSITIVE LOGITS

gni

0.27

ì²´

0.27

 opendir

0.25

dba

0.25

aviest

0.24

èĤīä½ĵ

0.24

æľĢåħĪ

0.24

 electronically

0.23

å±ĬæĹ¶

0.23

ierce

0.23

Activations Density 0.015%

news and academic articles

No Known Activations