INDEX

Explanations

programming-related terms and code snippets

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_scefr-ajt/6-res_scefr-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

 Hamm

-0.22

 understatement

-0.20

 neutrality

-0.19

ERN

-0.18

 hamm

-0.18

 Mand

-0.18

 McDonnell

-0.18

Bok

-0.18

 Rebirth

-0.17

Ham

-0.17

POSITIVE LOGITS

uled

0.23

wcs

0.22

 toget

0.21

eport

0.21

nown

0.21

atus

0.21

handled

0.21

hovah

0.21

 aven

0.20

icter

0.20

Activations Density 0.003%

programming-related terms and code snippets

No Comments

No Known Activations