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

 Shutterstock

-0.67

 thous

-0.67

xual

-0.67

 sober

-0.66

 Canaver

-0.64

BYU

-0.63

erald

-0.60

reflect

-0.60

Bye

-0.58

talk

-0.58

POSITIVE LOGITS

BuyableInstoreAndOnline

0.70

 assets

0.69

Os

0.64

MENTS

0.63

LEY

0.63

otal

0.63

imeters

0.62

Ot

0.61

ktop

0.61

otics

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.