INDEX

Explanations

references related to political figures and events

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

dayName

-0.94

Magikarp

-0.88

ãĤº

-0.81

cript

-0.77

<?

-0.76

ensing

-0.76

entric

-0.73

atana

-0.71

FTWARE

-0.70

RNA

-0.70

POSITIVE LOGITS

ufact

1.18

shire

0.85

ality

0.83

icia

0.82

 Yard

0.81

ridge

0.77

Osw

0.77

 Glas

0.77

ingham

0.75

 Blair

0.75

Activations Density 6.228%

references related to political figures and events

No Comments

No Known Activations