INDEX

Explanations

information related to technology and online platforms

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Examiner

-0.27

Wem

-0.25

 Frie

-0.24

icum

-0.24

 Pitch

-0.23

Ib

-0.23

 Mosque

-0.23

 Vive

-0.23

 Journalism

-0.23

 crow

-0.22

POSITIVE LOGITS

ĻĤ

0.32

ð

0.30

iddler

0.29

ł

0.28

isodes

0.28

ategories

0.28

ikes

0.27

¯

0.27

Â¯Â¯

0.26

Ĥª

0.26

Activations Density 10.872%

information related to technology and online platforms

No Comments

No Known Activations