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

à¸«à¸¥

-0.30

 sand

-0.30

æŁ¥éªĮ

-0.28

éŀĳ

-0.28

åĪ°è´¦

-0.27

unched

-0.27

 racial

-0.26

 contracting

-0.25

sand

-0.24

joined

-0.24

POSITIVE LOGITS

åĨĴ

0.32

åİŁå§ĭ

0.28

å¤ļäºĨ

0.27

åĬ³

0.27

èĩªçĦ¶

0.26

æĩ

0.26

enie

0.25

alia

0.25

prev

0.25

phen

0.24

Activations Density 0.002%

No Known Activations

This feature has no known activations.