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

çĲĥ

-0.28

balls

-0.27

imes

-0.26

older

-0.24

lug

-0.24

æĿµ

-0.24

erot

-0.24

IndexChanged

-0.23

enton

-0.23

å¯Į

-0.23

POSITIVE LOGITS

 clap

0.29

clid

0.24

 comment

0.24

 deciding

0.24

oras

0.24

rijk

0.24

 recognizing

0.23

è¿Ļä»¶äºĭ

0.23

 wash

0.23

ãĤ¯

0.23

Activations Density 0.017%

No Known Activations

This feature has no known activations.