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

soDeliveryDate

-0.95

DragonMagazine

-0.81

Mub

-0.80

 Debor

-0.79

Afee

-0.77

LOAD

-0.74

McA

-0.73

Wem

-0.73

LINE

-0.73

 Stain

-0.71

POSITIVE LOGITS

agara

0.90

£ı

0.83

ogly

0.83

icular

0.78

foreseen

0.77

ebus

0.76

erman

0.72

pmwiki

0.72

 overhead

0.71

uthor

0.71

Activations Density 0.000%

No Known Activations

This feature has no known activations.