INDEX

Explanations

Scientific publications and descriptions

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_23.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.23.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

zb

-0.30

(ec

-0.26

ĻĤ

-0.25

éĴ²

-0.25

ect

-0.25

ä¸ºæł¸å¿ĥçļĦ

-0.24

eries

-0.24

Peb

-0.24

çļĦè¯Ŀé¢ĺ

-0.24

mw

-0.23

POSITIVE LOGITS

ä¸įçĦ¶

0.28

edException

0.28

 addAction

0.27

Ð½ÑĮ

0.26

oning

0.26

Sik

0.26

ĲèĹı

0.26

erner

0.25

sss

0.25

ascus

0.25

Activations Density 0.005%

Scientific publications and descriptions

No Comments

No Known Activations