INDEX

Explanations

references to literature, history, and culture, particularly related to significant titles and characters

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

*/(

-0.83

espie

-0.83

oyer

-0.76

evin

-0.72

llah

-0.71

akeru

-0.71

 partName

-0.71

assetsadobe

-0.71

utherland

-0.71

UGC

-0.70

POSITIVE LOGITS

 Atlantis

0.87

 Omaha

0.82

 Souls

0.82

 Hearts

0.81

 Rochester

0.81

 sorts

0.80

 Rome

0.80

 whom

0.78

 course

0.78

Wak

0.78

Activations Density 10.123%

references to literature, history, and culture, particularly related to significant titles and characters

No Comments

No Known Activations