INDEX

Explanations

phrases related to social and political controversies or uprisings

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

é¾įå

-0.62

ACTED

-0.58

matched

-0.53

chens

-0.52

chester

-0.51

 afford

-0.50

 Paddock

-0.49

 undone

-0.48

 flats

-0.48

 OnePlus

-0.48

POSITIVE LOGITS

ire

0.75

 disbelief

0.72

 wrath

0.72

 controversy

0.71

 fury

0.70

 curiosity

0.69

 frenzy

0.69

 creativity

0.68

 indignation

0.67

 laughter

0.67

Activations Density 10.956%

phrases related to social and political controversies or uprisings

No Comments

No Known Activations