INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ilee

-0.72

 Directorate

-0.70

 Grave

-0.69

 Muse

-0.69

MOR

-0.63

PDF

-0.62

Haj

-0.61

ç·

-0.61

 Jinn

-0.60

Cah

-0.60

POSITIVE LOGITS

anti

0.84

ussia

0.76

uay

0.73

 Cosponsors

0.72

unct

0.69

independent

0.69

license

0.68

heast

0.67

odium

0.67

ntil

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.