INDEX

Explanations

proper nouns like names of individuals

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

stract

-0.70

ADRA

-0.70

ctory

-0.68

Ã§a

-0.66

UGE

-0.64

PDATE

-0.64

nces

-0.63

 sympt

-0.62

 pregn

-0.61

 Greenpeace

-0.61

POSITIVE LOGITS

sonian

1.58

son

0.93

smanship

0.90

anity

0.87

ies

0.86

tein

0.84

 Barney

0.84

inelli

0.83

sburg

0.83

gren

0.81

Activations Density 7.762%

proper nouns like names of individuals

No Comments

No Known Activations