INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 hypert

-0.60

 Kuala

-0.59

 Bottom

-0.59

Bey

-0.57

 standout

-0.57

 Adren

-0.57

 Graham

-0.56

 glut

-0.56

VID

-0.55

 phenomenal

-0.55

POSITIVE LOGITS

idav

0.76

illance

0.72

yrinth

0.71

rawdownloadcloneembedreportprint

0.70

quickShipAvailable

0.70

rection

0.69

bender

0.68

 Golem

0.68

0.67

artment

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.