INDEX

Explanations

Code

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 unins

-0.29

pot

-0.26

 ench

-0.25

Dock

-0.25

ociety

-0.24

èµ

-0.23

hec

-0.23

èµĦæºĲåħ±äº«

-0.23

nock

-0.23

 indiv

-0.23

POSITIVE LOGITS

åĲĪéĢĤ

0.28

åħĭæľį

0.25

amiliar

0.25

 sofort

0.24

æĸ¹éĿ¢

0.24

 IRequest

0.24

roscope

0.24

ç§°

0.23

Ã¤n

0.23

å¸¸è§Ħ

0.23

Activations Density 0.004%

Code

No Comments

No Known Activations