INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.32.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

åµ¬

-0.31

IONS

-0.27

ION

-0.27

aud

-0.27

åĬ²

-0.26

ioni

-0.26

ussion

-0.26

ione

-0.26

å¥Ķ

-0.25

POSITIVE LOGITS

anmar

0.29

 paraph

0.27

 deque

0.26

ppv

0.26

ellas

0.25

æ·ĳ

0.25

Mal

0.24

è¿¢

0.24

 Astroph

0.24

 tanks

0.23

Activations Density 0.000%

No Known Activations

This feature has no known activations.