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

é»İæĺİ

-0.27

æĥ³åĪ°

-0.27

trib

-0.26

ç¨½

-0.25

 pudd

-0.25

.orm

-0.25

 vanilla

-0.25

Õ¬

-0.24

åĽ½æľī

-0.24

 peeled

-0.24

POSITIVE LOGITS

imoto

0.28

æĹĥ

0.28

asin

0.26

 cough

0.26

anza

0.25

ÑĥÑħ

0.25

åĴ³åĹ½

0.25

æŃ¤æ¡Ī

0.24

ĪæĿĥ

0.23

duc

0.23

Activations Density 0.056%

No Known Activations

This feature has no known activations.