INDEX

Explanations

proper nouns, especially names of people

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

icity

-0.80

ingham

-0.73

ician

-0.70

iants

-0.63

 Beckham

-0.63

iets

-0.61

culosis

-0.60

urated

-0.60

antly

-0.60

ABE

-0.59

POSITIVE LOGITS

quez

0.76

leen

0.73

quel

0.71

§

0.71

uss

0.71

ven

0.67

©¶æ

0.67

emer

0.66

avior

0.66

axter

0.63

Activations Density 7.526%

proper nouns, especially names of people

No Comments

No Known Activations