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

ropolitan

-0.71

 Hour

-0.66

gency

-0.64

otes

-0.63

mile

-0.59

 Gram

-0.58

 Battery

-0.57

shire

-0.57

 MIDI

-0.56

Years

-0.56

POSITIVE LOGITS

orno

0.73

 corrid

0.71

apons

0.71

ppel

0.70

Ĥİ

0.70

xual

0.69

enges

0.69

 referen

0.69

Cool

0.67

WARE

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.