INDEX

Explanations

mentions of political figures and government positions, particularly Prime Ministers

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

MpServer

-0.81

 needles

-0.77

rights

-0.72

ĻĤ

-0.70

 fingert

-0.68

banks

-0.68

ntax

-0.67

wcsstore

-0.66

wd

-0.66

gew

-0.65

POSITIVE LOGITS

 Shinzo

1.02

 Narendra

0.98

 Theresa

0.86

 Stephen

0.84

 Ministers

0.79

 Minister

0.77

 Recep

0.76

 Tycoon

0.76

 Dmitry

0.75

 Benjamin

0.75

Activations Density 6.434%

mentions of political figures and government positions, particularly Prime Ministers

No Comments

No Known Activations