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

Not in Any Lists

No Comments

Negative Logits

ç¹ģ

-0.26

<![

-0.25

 penet

-0.25

å®£

-0.25

integr

-0.24

çķª

-0.24

å®ĭ

-0.23

Le

-0.23

 lokal

-0.23

çªł

-0.23

POSITIVE LOGITS

æĭ¾

0.25

æ¶ī

0.25

TRACT

0.25

ĺìĿ´

0.24

ewood

0.24

 thanked

0.24

æĤ¨çļĦåŃ©åŃĲ

0.24

fal

0.24

amil

0.23

å¯°

0.23

Activations Density 0.020%

No Known Activations

This feature has no known activations.