INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_19.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.19.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

èĥ½ä¸įèĥ½

-0.31

 attained

-0.30

æİĪè¯¾

-0.28

Whether

-0.26

MG

-0.26

equip

-0.26

èĥ½åĲ¦

-0.25

 Whether

-0.25

 perc

-0.25

-current

-0.24

POSITIVE LOGITS

nid

0.29

ä»İä¸Ń

0.27

 brows

0.26

ismet

0.25

naires

0.25

gien

0.25

tablet

0.25

lder

0.25

oppable

0.24

urous

0.24

Activations Density 6.655%

No Known Activations

This feature has no known activations.