INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_20.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.20.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

stalk

-0.29

ä¸Ģçº§

-0.27

éħį

-0.27

oman

-0.27

å¥½çľĭ

-0.25

ä¼Ļä¼´åħ³ç³»

-0.25

å½ĵä¹ĭ

-0.24

sca

-0.24

éħįä»¶

-0.24

?):

-0.24

POSITIVE LOGITS

 negatives

0.26

ä¸Ńä»ĭ

0.25

yclopedia

0.25

èĢģäººå®¶

0.25

 Nullable

0.25

Ð¾Ð±Ð¾ÑĢÐ¾ÑĤ

0.24

amera

0.24

 studio

0.24

 WikiLeaks

0.24

erais

0.24

Activations Density 4.916%

No Known Activations

This feature has no known activations.