INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

sie

-0.71

 Coffin

-0.66

arnaev

-0.66

finder

-0.65

assetsadobe

-0.64

Patch

-0.64

mart

-0.64

utsch

-0.63

NSA

-0.60

bill

-0.60

POSITIVE LOGITS

dri

0.75

ifiable

0.72

Pwr

0.71

erved

0.70

resents

0.69

ÃŃs

0.68

oubted

0.67

IFIED

0.65

 coasts

0.65

Í

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.