INDEX

Explanations

vector-related terms in technical instructions and data sets

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

doms

-0.72

soDeliveryDate

-0.65

ecause

-0.65

perty

-0.63

kens

-0.61

rahim

-0.61

rontal

-0.61

ilial

-0.61

scrib

-0.60

lished

-0.60

POSITIVE LOGITS

 Vector

0.74

vector

0.70

 vector

0.69

ization

0.62

izable

0.62

ocity

0.60

idian

0.60

iz

0.59

icular

0.58

ized

0.58

Activations Density 6.771%

vector-related terms in technical instructions and data sets

No Comments

No Known Activations