INDEX

Explanations

is

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_31.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.31.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Cle

-0.32

clear

-0.30

æ¸ħ

-0.29

OK

-0.29

cle

-0.29

dark

-0.28

_OK

-0.27

 Rein

-0.26

	clear

-0.26

 Clear

-0.26

POSITIVE LOGITS

çļĦåŃ¦ä¹ł

0.28

æªĶ

0.27

è¿ģ

0.26

anted

0.26

ASE

0.24

äºĮçº§

0.24

/bower

0.24

 tern

0.24

çļĦæĢ§æł¼

0.24

å¦Ĥä½ķåİ»

0.24

Activations Density 0.021%

is

No Comments

No Known Activations