INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_20.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.20.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Ð¾Ð²Ð°ÑĢ

-0.26

 Tire

-0.25

 footing

-0.24

CLU

-0.24

.ravel

-0.23

udded

-0.23

.inline

-0.23

odate

-0.23

æĤ¬å´ĸ

-0.23

IRTUAL

-0.23

POSITIVE LOGITS

 compens

0.28

 fright

0.27

aille

0.26

ä¼¤

0.26

 cath

0.25

 bless

0.25

 hurt

0.25

 injuries

0.24

hyp

0.24

 bias

0.23

Activations Density 2.055%

No Known Activations

This feature has no known activations.