INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

auri

-0.71

allo

-0.70

uran

-0.68

xon

-0.68

efully

-0.67

spir

-0.66

esville

-0.66

bite

-0.64

ature

-0.63

eworks

-0.62

POSITIVE LOGITS

WB

0.70

LOAD

0.69

Beg

0.63

çİĭ

0.62

 captcha

0.62

 è£ıè

0.62

CRIP

0.61

UFF

0.61

EE

0.60

 RandomRedditorWithNo

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.