INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_20.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.20.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

ÅĽci

-0.30

yah

-0.28

iah

-0.27

æĪĲå°±æĦŁ

-0.26

ulan

-0.25

çĤ¹éĴŁ

-0.25

unh

-0.24

ÙĬÙĩ

-0.24

lige

-0.24

:req

-0.24

POSITIVE LOGITS

//=

0.26

tas

0.26

DDS

0.25

izations

0.24

://

0.24

tg

0.24

vard

0.24

AZE

0.24

DJ

0.24

//@

0.24

Activations Density 17.544%

No Known Activations

This feature has no known activations.