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

point

-0.27

ç¬¼

-0.26

é¡¶

-0.26

æĥ³ä¸įåĪ°

-0.26

append

-0.26

ghost

-0.25

supply

-0.25

 balk

-0.25

é¸¦

-0.25

æī©å±ķ

-0.25

POSITIVE LOGITS

ä¿Ń

0.30

_Save

0.28

ssel

0.27

 Preservation

0.25

 saving

0.25

 preservation

0.25

 visually

0.24

 Prepare

0.24

çį¨

0.24

 Preparation

0.24

Activations Density 79.921%

No Known Activations

This feature has no known activations.