INDEX

Explanations

questions and discussions related to technology and policy-making

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

 Skull

-0.74

DragonMagazine

-0.74

éĹĺ

-0.73

Reloaded

-0.72

 Ridley

-0.72

 Strawberry

-0.71

åį

-0.70

 Monk

-0.68

Kiw

-0.68

ilts

-0.68

POSITIVE LOGITS

po

0.93

di

0.92

selves

0.89

la

0.88

nesota

0.87

ten

0.83

go

0.82

late

0.80

fi

0.80

dra

0.80

Activations Density 5.204%

questions and discussions related to technology and policy-making

No Comments

No Known Activations