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

yang

-0.77

IRC

-0.75

."[

-0.72

 Â¥

-0.70

Ð½

-0.67

 includ

-0.65

Iro

-0.65

 Okinawa

-0.65

".[

-0.65

 Japan

-0.65

POSITIVE LOGITS

 ticking

0.80

mble

0.73

DragonMagazine

0.73

£ı

0.72

emort

0.72

ļéĨĴ

0.72

 playback

0.70

pod

0.70

rians

0.67

¥µ

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.