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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

omaly

-0.80

catch

-0.68

ifles

-0.64

QR

-0.63

duino

-0.60

zzle

-0.59

ulia

-0.58

keep

-0.57

anol

-0.57

ocr

-0.57

POSITIVE LOGITS

gress

0.72

 skelet

0.70

ishers

0.69

aunted

0.68

 Frie

0.68

escription

0.66

CLUS

0.65

à¥

0.65

Winged

0.65

isher

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.