INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

NetMessage

-0.71

»Ĵ

-0.69

 footing

-0.68

okÃ©

-0.68

 guild

-0.67

PokÃ©mon

-0.66

oxy

-0.66

AY

-0.66

ĨĴ

-0.64

ay

-0.64

POSITIVE LOGITS

ruary

0.90

enegger

0.67

 Conversion

0.65

 Hobby

0.65

ials

0.65

ieties

0.64

arette

0.63

less

0.63

iability

0.62

Fax

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.