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

 Park

-0.29

-0.27

äººåľ¨

-0.27

Ãī

-0.26

 Elect

-0.25

Art

-0.25

oment

-0.24

Elect

-0.24

èµ°å»Ĭ

-0.24

 Chow

-0.23

POSITIVE LOGITS

 conservatism

0.27

 granny

0.26

ounc

0.25

ä¸»ä½ĵè´£ä»»

0.25

:\"

0.24

otec

0.24

acidad

0.24

 burned

0.24

 Grinder

0.24

å®īåħ¨ä¿Ŀéļľ

0.24

Activations Density 0.006%

No Known Activations

This feature has no known activations.