INDEX

Explanations

Roman numerals

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_22.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.22.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

edException

-0.30

standing

-0.25

nable

-0.24

sort

-0.23

nels

-0.23

åŃµåĮĸåĻ¨

-0.23

 bulund

-0.23

çľĭä¸įæĩĤ

-0.22

nost

-0.22

ç§¦çļĩ

-0.22

POSITIVE LOGITS

ieme

0.27

peat

0.26

åł°

0.25

 Fleet

0.24

(!!

0.24

abetes

0.23

è¯Ŀ

0.23

 Ð²Ð°Ð»

0.23

ungal

0.23

äº¥

0.23

Activations Density 0.246%

Roman numerals

No Known Activations