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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ç®¬

-0.28

éĻĦä»¶

-0.27

emarks

-0.27

anmar

-0.25

BackingField

-0.25

 basket

-0.24

burger

-0.24

basket

-0.23

oland

-0.23

lÃ¢

-0.23

POSITIVE LOGITS

å¥

0.26

ç©ºæ°Ķ

0.25

 thÄĥm

0.25

å»ºåĽ½

0.24

 airflow

0.24

_SCOPE

0.24

ears

0.24

air

0.23

çļĦæĹ¶éĹ´éĩĮ

0.23

 snatch

0.23

Activations Density 0.008%

No Known Activations

This feature has no known activations.