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

 burgl

-0.78

 Protector

-0.67

rape

-0.66

falls

-0.65

CVE

-0.64

raz

-0.64

 recess

-0.63

 Revelations

-0.63

 Bounty

-0.62

essions

-0.62

POSITIVE LOGITS

esm

0.73

çĲ

0.73

Äĩ

0.71

åŃ

0.70

¢

0.69

thumbnails

0.68

achus

0.68

ãĤ°

0.68

anguage

0.67

ikhail

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.