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

ä½¿äºº

-0.28

ä½¿

-0.28

 Axis

-0.27

èĩªçĶ±

-0.27

 Dere

-0.26

mani

-0.25

.outer

-0.25

å¤§çĲĨçŁ³

-0.25

-setup

-0.25

ç»¨

-0.25

POSITIVE LOGITS

åĴ§

0.28

å«Ĥ

0.27

æĴ¸

0.25

èĭĹ

0.25

 spel

0.24

rons

0.24

lfw

0.24

é¡ºä¾¿

0.24

GY

0.23

riÃ³

0.23

Activations Density 7.206%

No Known Activations

This feature has no known activations.