INDEX

Explanations

verbal phrases related to actions and events

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

ãĥĥãĥī

-0.90

 encl

-0.90

BIT

-0.89

 corrid

-0.83

 cath

-0.82

²¾

-0.82

iHUD

-0.80

âĢ¢âĢ¢âĢ¢âĢ¢

-0.80

ushima

-0.80

gemony

-0.78

POSITIVE LOGITS

dates

1.95

stairs

1.54

dating

1.44

olean

1.26

rison

1.21

grade

1.14

rons

1.13

olicy

1.13

oons

1.11

graded

1.08

Activations Density 0.774%

verbal phrases related to actions and events

No Comments

No Known Activations