INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_19.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.19.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

è°²

-0.26

mart

-0.25

hall

-0.25

è¿Ĥ

-0.24

åİļ

-0.24

ieurs

-0.23

trees

-0.23

 sheets

-0.23

æŀ¶

-0.23

heets

-0.23

POSITIVE LOGITS

vides

0.29

æŃī

0.28

 distract

0.26

å¼ºè¡Į

0.26

ä¹ŀ

0.25

 offended

0.25

kses

0.25

ä¸įæĽ¾

0.24

SSIP

0.24

fell

0.24

Activations Density 1.671%

No Known Activations

This feature has no known activations.