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

.Any

-0.27

osed

-0.27

 technological

-0.26

-gap

-0.25

 morning

-0.24

æ¾¹

-0.24

ï¼ģãĢį

-0.24

ORED

-0.24

 ....↵↵

-0.23

 straightforward

-0.23

POSITIVE LOGITS

æĿĸ

0.31

 Beans

0.26

cade

0.26

.bundle

0.25

oire

0.25

(@(

0.24

itaire

0.24

 dint

0.24

interactive

0.24

restriction

0.24

Activations Density 0.002%

No Known Activations

This feature has no known activations.