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

Not in Any Lists

No Comments

Negative Logits

æ°ĳ

-0.29

OPS

-0.28

lian

-0.28

Opr

-0.28

æ¯«

-0.28

åħ¨æ°ĳ

-0.26

opi

-0.26

arians

-0.26

å¤ĸæĿ¥

-0.25

lic

-0.25

POSITIVE LOGITS

 servi

0.28

ekte

0.24

åĿļå®ļä¸įç§»

0.24

cuts

0.24

_STRUCT

0.24

 struck

0.24

uida

0.24

 ship

0.23

chner

0.23

 Struct

0.23

Activations Density 2.503%

No Known Activations

This feature has no known activations.