INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

unia

-0.76

iral

-0.74

ificate

-0.73

 tradem

-0.73

EStream

-0.71

 mosqu

-0.70

icides

-0.70

utterstock

-0.68

ngth

-0.67

awaru

-0.65

POSITIVE LOGITS

rider

0.92

 Brist

0.88

TW

0.80

BRE

0.71

bre

0.70

BU

0.70

Israel

0.69

NAT

0.69

ECA

0.69

India

0.68

Activations Density 0.000%

No Known Activations

This feature has no known activations.