INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_18.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.18.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

oire

-0.25

adol

-0.25

 pride

-0.25

.fig

-0.25

éĩįè¦ģæĦıä¹ī

-0.25

çħ§æł·

-0.24

thane

-0.24

lus

-0.24

 Guards

-0.24

');?></

-0.24

POSITIVE LOGITS

eds

0.30

éĻµ

0.30

åĪĨæķ£

0.28

æıĲçĤ¼

0.25

olars

0.24

urator

0.24

rng

0.24

 dispersed

0.24

 scattered

0.23

pez

0.23

Activations Density 33.100%

No Known Activations

This feature has no known activations.