INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

çĶ°

-0.80

assetsadobe

-0.77

schild

-0.76

IVERS

-0.72

earable

-0.72

etsk

-0.72

xual

-0.72

Ĥİ

-0.72

IDENT

-0.70

VK

-0.70

POSITIVE LOGITS

cat

0.71

breaker

0.70

cap

0.65

eers

0.64

bird

0.62

 Atlantis

0.62

teenth

0.61

cellent

0.61

esthes

0.61

 Squid

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.