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

 Copenhagen

-0.74

feature

-0.70

xes

-0.69

 Europeans

-0.68

interstitial

-0.68

 Danish

-0.65

 Travels

-0.63

arget

-0.63

crop

-0.62

anwhile

-0.62

POSITIVE LOGITS

Jr

0.72

 Guard

0.65

Contents

0.64

 uncont

0.64

yon

0.63

uer

0.63

tarian

0.63

unia

0.60

gt

0.60

 tears

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.