INDEX

Explanations

references to software packages or modules

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

nesota

-0.78

alties

-0.68

bered

-0.65

aft

-0.64

shire

-0.61

ffen

-0.59

baugh

-0.58

farious

-0.57

izabeth

-0.56

auga

-0.55

POSITIVE LOGITS

 Contents

0.73

packages

0.72

 packages

0.70

 Package

0.64

 package

0.61

 containing

0.60

 Includes

0.59

 manifests

0.58

nam

0.57

aging

0.57

Activations Density 10.694%

references to software packages or modules

No Comments

No Known Activations