INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

PHOTOS

-0.80

 Vigil

-0.79

iscopal

-0.75

>:

-0.69

rina

-0.69

Â·Â·

-0.67

Enlarge

-0.67

aido

-0.67

SPONSORED

-0.67

EAR

-0.66

POSITIVE LOGITS

aber

0.76

creen

0.68

ubby

0.67

 painting

0.66

 chall

0.64

 selling

0.63

 seller

0.63

 reckoned

0.62

seller

0.62

apon

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.