INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_1.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.1.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

ities

-0.27

 Sark

-0.27

otten

-0.27

ãĥĲãĥ³

-0.25

zens

-0.25

ä¸»åħ¬

-0.25

/tab

-0.25

onis

-0.25

åĲĮæĻĤãģ«

-0.24

umeric

-0.24

POSITIVE LOGITS

å¸Ń

0.30

pile

0.27

.hr

0.27

 hydr

0.27

 former

0.26

æµĵåİļ

0.26

hydr

0.24

é¦Ħ

0.24

åħľ

0.24

 entrev

0.24

Activations Density 0.000%

No Known Activations

This feature has no known activations.