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

atre

-0.78

 Compare

-0.69

bard

-0.68

witz

-0.65

 REPORT

-0.65

orman

-0.65

ivals

-0.64

advertising

-0.64

 Papers

-0.63

VILLE

-0.62

POSITIVE LOGITS

pless

0.70

ulous

0.68

uly

0.66

uts

0.64

esides

0.63

atted

0.63

ilty

0.63

ulously

0.60

kh

0.60

angled

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.