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

arl

-0.29

(optimizer

-0.27

ä¼ĳ

-0.26

Gow

-0.26

GenerationStrategy

-0.25

 DeepCopy

-0.24

ress

-0.24

å®ŀè¡Į

-0.24

@param

-0.24

ÙĦÙĥ

-0.24

POSITIVE LOGITS

éĤ£æ¬¡

0.29

éĩį

0.27

æ¶Ķ

0.26

çªŁ

0.25

_here

0.25

è¿Ļä¸īä¸ª

0.24

æ¯ıæ¬¡

0.24

è¿Ļæī¹

0.24

-toggler

0.24

ioc

0.24

Activations Density 2.140%

No Known Activations

This feature has no known activations.