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

anship

-0.81

omal

-0.77

iller

-0.76

aughtered

-0.76

geoning

-0.76

omial

-0.74

pter

-0.71

ulic

-0.71

hov

-0.70

atial

-0.70

POSITIVE LOGITS

bah

0.66

 printing

0.66

idas

0.64

 printers

0.62

 unprepared

0.61

 debugging

0.59

0.58

 printer

0.57

 luggage

0.57

 Horowitz

0.56

Activations Density 0.000%

No Known Activations

This feature has no known activations.