INDEX

Explanations

No Explanations Found

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

å¹Į

-0.30

åľĪ

-0.27

enity

-0.27

Ð³Ð¾ÑĢÐ¾Ð´

-0.26

urrences

-0.26

äº²

-0.25

çīĮ

-0.25

aters

-0.24

_tweets

-0.23

æŃĩ

-0.23

POSITIVE LOGITS

èĥĮä¸Ĭ

0.29

 dispenser

0.27

IDM

0.25

ä¸Ńæĸ¹

0.24

 mole

0.24

 shortcut

0.23

.IsValid

0.23

 ints

0.23

è²Ĭ

0.23

-back

0.23

Activations Density 0.002%

No Known Activations

This feature has no known activations.