INDEX

Explanations

numerical references to specific chapters within a book or series

oai_token-act-pair · gpt-3.5-turbo

references to specific chapters in a book or document

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

oggles

-0.70

berman

-0.69

zzle

-0.67

fitting

-0.66

 blaster

-0.65

enez

-0.64

yg

-0.64

pes

-0.63

 fitt

-0.63

vant

-0.62

POSITIVE LOGITS

ĸļ

1.15

 chapter

0.83

acters

0.82

 chapters

0.82

 Chapters

0.80

icularly

0.78

chapter

0.76

naire

0.76

book

0.74

hound

0.73

Activations Density 0.019%

numerical references to specific chapters within a book or series

references to specific chapters in a book or document

No Comments

No Known Activations