INDEX

Explanations

double

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ibs

-0.26

Ä±yla

-0.25

uci

-0.25

ä¾Ŀæ¬¡

-0.24

è¾ĩ

-0.24

 Chevy

-0.24

ÑĪÐºÐ¸

-0.23

è¼

-0.23

 successor

-0.22

AGED

-0.22

POSITIVE LOGITS

åıĮ

1.38

 double

1.30

åıĮéĩį

1.23

 dual

1.18

double

1.16

éĽĻ

1.12

 Double

1.06

Double

1.03

 Dual

1.01

-double

1.00

Activations Density 0.007%

double

No Comments

No Known Activations