INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_31.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.31.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

çŀ½

-0.26

 Libre

-0.25

/stretch

-0.23

eing

-0.23

apia

-0.23

.yang

-0.22

nan

-0.22

tractor

-0.22

andro

-0.22

apolis

-0.22

POSITIVE LOGITS

IGN

0.26

éļ¶å±ŀ

0.26

 hail

0.25

éĩįå¤į

0.25

oves

0.25

æĲģ

0.24

åĨ²åĬ¨

0.24

ICE

0.23

qrst

0.23

å°ıæĹ¶åīį

0.22

Activations Density 0.002%

No Known Activations

This feature has no known activations.