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

rawdownloadcloneembedreportprint

-0.75

quartered

-0.73

irs

-0.72

rongh

-0.72

zzle

-0.71

eki

-0.71

soDeliveryDate

-0.70

 handc

-0.69

ires

-0.69

£ı

-0.69

POSITIVE LOGITS

 Designer

0.69

 rave

0.68

 Discovery

0.67

 Creative

0.67

Nic

0.65

Justin

0.64

making

0.64

 Rampage

0.63

AME

0.62

 Pablo

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.