INDEX

Explanations

warnings about graphic content or sensitive material in text data

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

issance

-0.87

obbies

-0.70

along

-0.68

externalActionCode

-0.65

pled

-0.65

kefeller

-0.64

iscover

-0.64

inguished

-0.64

riers

-0.63

assadors

-0.62

POSITIVE LOGITS

 spoilers

1.42

OIL

1.25

WARNING

1.19

 Contains

1.19

 spoiler

1.11

 IMAGES

1.07

 Spoiler

0.99

 nudity

0.97

Warning

0.94

 graphic

0.93

Activations Density 0.193%

warnings about graphic content or sensitive material in text data

No Comments

No Known Activations