INDEX

Explanations

references to historical and religious events or figures

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-ajt

Prompts (Dashboard)

12,288 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

46,080

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

apollo-research/Skylion007-openwebtext-tokenizer-gpt2

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

enegger

-1.09

 hobbies

-1.00

 actionGroup

-0.99

ockey

-0.91

 guiIcon

-0.91

gain

-0.88

ronics

-0.88

assetsadobe

-0.88

addons

-0.88

projects

-0.87

POSITIVE LOGITS

 Hebrew

1.45

 Exodus

1.43

 Psal

1.37

iblical

1.32

 Ezekiel

1.28

 Scripture

1.27

 Romans

1.26

 Neph

1.26

á¸

1.25

ospels

1.24

Activations Density 10.616%

references to historical and religious events or figures

No Comments

No Known Activations