INDEX

Explanations

mentions of or related to terrorist activities

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

galitarian

-0.91

shire

-0.76

laus

-0.74

)=(

-0.72

BuyableInstoreAndOnline

-0.71

dit

-0.71

rose

-0.68

ðĿ

-0.67

aver

-0.65

iband

-0.64

POSITIVE LOGITS

abad

1.08

 mastermind

0.96

 bombings

0.95

 bombing

0.91

 terrorist

0.88

 spree

0.85

 attacks

0.84

 terrorists

0.83

 bombers

0.83

 terrorism

0.81

Activations Density 11.060%

mentions of or related to terrorist activities

No Comments

No Known Activations