INDEX

Explanations

technical terms and phrases related to intelligence work, technology, and military operations

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

igs

-0.74

ascript

-0.71

xia

-0.71

agos

-0.70

igate

-0.70

ieval

-0.68

BuyableInstoreAndOnline

-0.67

YP

-0.64

oreal

-0.63

imental

-0.62

POSITIVE LOGITS

 therein

0.82

cru

0.78

 importantly

0.77

 horm

0.76

 thereafter

0.72

 damned

0.71

 downright

0.69

 hilar

0.69

 consequently

0.69

 creep

0.68

Activations Density 18.136%

technical terms and phrases related to intelligence work, technology, and military operations

No Comments

No Known Activations