INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_22.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.22.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

inia

-0.27

outh

-0.26

 corro

-0.26

 difference

-0.25

ä»£çĲĨäºº

-0.24

ç¿ķ

-0.24

ibility

-0.24

ron

-0.23

çİĩä¸º

-0.23

kul

-0.23

POSITIVE LOGITS

Ã¼nde

0.31

erland

0.28

 clip

0.27

-INF

0.27

atsu

0.26

_allocate

0.26

 Clip

0.24

-income

0.24

ukt

0.24

odied

0.23

Activations Density 0.000%

No Known Activations

This feature has no known activations.