INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_23.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.23.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

bench

-0.30

æĮĳæĪĺ

-0.28

ron

-0.27

 shutter

-0.26

us

-0.24

 disturb

-0.24

RON

-0.24

ops

-0.24

LM

-0.24

 ecology

-0.24

POSITIVE LOGITS

ä¹ĭæīĢ

0.25

å¾Ĺèµ·

0.25

 strr

0.24

essen

0.24

SYN

0.24

 comment

0.24

 Alexand

0.24

agine

0.24

åī©

0.24

æĮĿ

0.24

Activations Density 0.007%

No Known Activations

This feature has no known activations.