INDEX

Explanations

information related to news, articles, and regulations

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

enance

-0.52

lee

-0.52

ongevity

-0.50

IONS

-0.50

 brim

-0.49

igree

-0.48

endi

-0.48

OVER

-0.48

ikan

-0.46

 Winning

-0.46

POSITIVE LOGITS

ÄŁ

0.54

 WATCHED

0.54

TBA

0.53

ebin

0.52

 Annotations

0.52

 decrypt

0.51

 millenn

0.50

 installments

0.50

ulkan

0.50

 drafts

0.50

Activations Density 2.571%

information related to news, articles, and regulations

No Comments

No Known Activations