INDEX

Explanations

phrases related to physical or mental effort

oai_token-act-pair · gpt-3.5-turbo

references to effort or labor

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

 Ukrain

-0.87

 constitu

-0.81

wcs

-0.81

anamo

-0.76

iren

-0.73

 Flavoring

-0.72

 sidx

-0.70

EStream

-0.69

 Bubble

-0.68

netflix

-0.66

POSITIVE LOGITS

 ethic

1.37

station

1.26

flows

1.25

manship

1.21

horse

1.16

bench

1.15

aday

1.11

fare

1.06

forces

0.98

work

0.94

Activations Density 0.063%

phrases related to physical or mental effort

references to effort or labor

No Comments

No Known Activations