INDEX

Explanations

phrases related to investigations, researching, and journalism

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Fle

-0.70

Spoiler

-0.67

externalActionCode

-0.67

BR

-0.66

advertisement

-0.65

cially

-0.65

Netflix

-0.64

Null

-0.64

Pinterest

-0.63

)</

-0.63

POSITIVE LOGITS

 Souls

0.68

 Mile

0.64

 Travels

0.63

 Pyramid

0.58

Sut

0.58

 Agent

0.55

ierrez

0.55

wagen

0.55

 Slim

0.53

 specializing

0.53

Activations Density 0.028%

phrases related to investigations, researching, and journalism

No Comments

No Known Activations