INDEX

Explanations

variables and coding terms related to computer programming

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

assetsadobe

-0.64

 behav

-0.61

 colle

-0.60

 Vide

-0.57

iP

-0.57

voc

-0.56

 financially

-0.56

 persist

-0.56

 compet

-0.55

 recomm

-0.55

POSITIVE LOGITS

zon

0.82

ph

0.79

roth

0.78

roy

0.78

ox

0.76

gain

0.75

posed

0.74

po

0.71

mere

0.71

fu

0.71

Activations Density 22.480%

variables and coding terms related to computer programming

No Comments

No Known Activations