INDEX

Explanations

mentions of specific people, organizations, and locations related to political or controversial topics

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

 straps

-0.69

 ceilings

-0.68

 signatures

-0.66

 layouts

-0.65

 Able

-0.64

 Resp

-0.64

 cables

-0.64

 certificates

-0.63

 lashes

-0.62

 endorsement

-0.62

POSITIVE LOGITS

arma

0.93

ellectual

0.89

osph

0.88

oton

0.83

etheless

0.79

roleum

0.78

onymous

0.78

nered

0.78

sterdam

0.78

apest

0.77

Activations Density 8.007%

mentions of specific people, organizations, and locations related to political or controversial topics

No Comments

No Known Activations