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

 drip

-0.26

 extractor

-0.25

ones

-0.24

Compact

-0.24

 examiner

-0.24

tes

-0.24

 Starr

-0.23

 Extraction

-0.23

upe

-0.23

 Consultants

-0.23

POSITIVE LOGITS

 pockets

0.29

éĹ³

0.26

 sensit

0.25

Ð¾ÑģÑĤÐµÐ¹

0.25

 tinder

0.25

ä¸¥èĤĥ

0.25

 initiator

0.24

-pocket

0.24

 nearby

0.24

 flare

0.24

Activations Density 0.003%

No Known Activations

This feature has no known activations.