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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 correspondence

-0.29

ities

-0.26

lap

-0.25

ç»¿èī²éĢļéģĵ

-0.25

okus

-0.24

æĭ³å¤´

-0.24

 hasher

-0.24

æĤ¨çļĦåŃ©åŃĲ

-0.24

èĨ³

-0.23

èµĦæºĲåħ±äº«

-0.23

POSITIVE LOGITS

äººç¾¤

0.26

EDIUM

0.25

åĪĿ

0.25

jdk

0.25

 vulnerable

0.25

-Jul

0.25

lems

0.24

 ÑģÐ²Ð¾Ð¸Ñħ

0.24

æ³Ħæ¼ı

0.24

SCO

0.24

Activations Density 0.006%

No Known Activations

This feature has no known activations.