INDEX

Explanations

phrases related to activities and actions

oai_token-act-pair · gpt-3.5-turbo

punctuation marks and their frequency within the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

bilt

-0.79

é¾įåĸļå£«

-0.76

 neighb

-0.68

undred

-0.67

aeda

-0.66

ucci

-0.66

azeera

-0.66

ãĤº

-0.64

å§«

-0.63

imate

-0.63

POSITIVE LOGITS

etc

0.87

aka

0.69

etc

0.68

 Shutterstock

0.68

eh

0.65

 accompanies

0.60

 courtesy

0.60

 Scientology

0.60

 thence

0.59

————————

0.59

Activations Density 1.145%

phrases related to activities and actions

punctuation marks and their frequency within the text

No Comments

No Known Activations