INDEX

Explanations

references to specific people's names or titles

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

CBI

-0.90

 toile

-0.82

[*]

-0.80

 mantle

-0.75

 Vaugh

-0.73

 Deadpool

-0.72

 Notting

-0.71

 Firm

-0.71

[+]

-0.69

 questioning

-0.68

POSITIVE LOGITS

waukee

1.18

gage

1.07

nih

1.02

intosh

1.01

ICAN

0.98

htar

0.97

ogi

0.95

arella

0.93

udi

0.92

ufact

0.88

Activations Density 16.011%

references to specific people's names or titles

No Comments

No Known Activations