INDEX

Explanations

phrases related to working, tasks, and collaboration

oai_token-act-pair · gpt-3.5-turbo

references to the concept of working or labor

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

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ylon

-0.80

Sov

-0.68

IX

-0.66

 Ukrain

-0.66

 Flavoring

-0.66

angular

-0.65

 Austral

-0.65

 constitu

-0.65

 Bubble

-0.64

Sov

-0.64

POSITIVE LOGITS

 ethic

1.21

bench

1.10

hops

0.99

manship

0.98

flows

0.94

station

0.89

mates

0.86

mate

0.85

aday

0.84

heet

0.84

Activations Density 0.091%

phrases related to working, tasks, and collaboration

references to the concept of working or labor

No Comments

No Known Activations