INDEX

Explanations

phrases related to legal terminologies and concepts, such as immunity, prosecution, and resistance

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

oric

-0.83

erker

-0.75

furt

-0.73

orical

-0.72

colo

-0.71

driver

-0.70

urgical

-0.70

straight

-0.69

geist

-0.69

yard

-0.68

POSITIVE LOGITS

 immunity

1.15

 afforded

0.92

ionage

0.75

 lapse

0.75

 pard

0.73

 pardon

0.73

 Immunity

0.71

 granted

0.69

awaru

0.68

 bask

0.67

Activations Density 6.656%

phrases related to legal terminologies and concepts, such as immunity, prosecution, and resistance

No Comments

No Known Activations