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

cxx

-0.31

aviest

-0.31

 Wheels

-0.27

CTest

-0.27

cription

-0.27

ptest

-0.26

plib

-0.26

éĤ£è¾¹

-0.25

Vectorizer

-0.25

ĵĺ

-0.25

POSITIVE LOGITS

 world

0.28

 claims

0.27

 just

0.27

 pure

0.26

åľ¨æķ´ä¸ª

0.26

 internal

0.26

æĢ»ä½ĵ

0.26

 floor

0.26

 dÃ©c

0.26

0.25

Activations Density 5.767%

No Known Activations

This feature has no known activations.