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

gim

-0.28

 mend

-0.25

NSS

-0.25

-0.24

è¯ķ

-0.23

.appendTo

-0.23

idal

-0.23

inan

-0.23

gv

-0.23

 Wass

-0.23

POSITIVE LOGITS

{:

0.28

urt

0.26

raction

0.26

alcon

0.26

Con

0.25

éĽĨè£ħ

0.25

Ø§ÙĩØ±

0.25

ä¸įæħİ

0.25

ä¸įå°ıå¿ĥ

0.25

éĦĦ

0.24

Activations Density 0.004%

No Known Activations

This feature has no known activations.