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

Not in Any Lists

No Comments

Negative Logits

åįģéĩĮ

-0.27

fin

-0.26

èĢĥ

-0.25

 NÃ¤

-0.24

åħĦå¼Ł

-0.24

erval

-0.24

isu

-0.23

 máº«u

-0.23

ä¸¤çĤ¹

-0.23

 Champion

-0.23

POSITIVE LOGITS

*</

0.27

Cli

0.26

++$

0.26

(stage

0.24

Sag

0.24

 evacuated

0.24

cern

0.23

+w

0.23

 Yahoo

0.23

oug

0.23

Activations Density 65.845%

No Known Activations

This feature has no known activations.