INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_19.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.19.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 premises

-0.28

 None

-0.26

Snippet

-0.25

AccessException

-0.25

çīĪ

-0.24

alt

-0.23

 Prem

-0.23

 Conce

-0.23

ayne

-0.23

templates

-0.23

POSITIVE LOGITS

oze

0.29

celed

0.28

åĿİ

0.27

 Coun

0.26

å¹¹

0.26

oster

0.26

lopedia

0.25

èµ°ä¸ĭåİ»

0.25

 catalog

0.24

catalog

0.24

Activations Density 0.016%

No Known Activations

This feature has no known activations.