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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

æ©Ł

-0.69

 Goddess

-0.68

Dad

-0.68

Gri

-0.68

 Enlight

-0.68

 Underground

-0.67

 Clicker

-0.67

 Behind

-0.67

way

-0.64

 Collider

-0.63

POSITIVE LOGITS

izon

0.84

interstitial

0.78

VIDIA

0.73

ivably

0.70

glers

0.69

ocy

0.68

inos

0.68

asma

0.68

imester

0.68

 orphans

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.