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

istar

-0.82

stals

-0.79

onder

-0.78

achus

-0.73

omore

-0.71

ometry

-0.71

anqu

-0.70

atinum

-0.69

SAM

-0.67

xus

-0.63

POSITIVE LOGITS

advertising

0.75

Decre

0.73

Church

0.72

SPONSORED

0.64

ouse

0.63

arial

0.63

ERY

0.62

iffe

0.62

Journal

0.61

Commission

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.