INDEX

Explanations

phrases related to law enforcement and crime investigations

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

çīĪ

-0.81

 Films

-0.74

 vows

-0.69

æ©

-0.69

pter

-0.69

ORGE

-0.63

 Fant

-0.63

 Advice

-0.62

*/(

-0.62

éĥ

-0.62

POSITIVE LOGITS

icably

0.97

 bother

0.96

 adequately

0.95

 properly

0.92

 bothering

0.89

 sufficiently

0.86

 adversely

0.85

 detectable

0.85

 exist

0.83

icable

0.83

Activations Density 4.555%

phrases related to law enforcement and crime investigations

No Comments

No Known Activations