INDEX

Explanations

mentions of political affiliations, specifically focusing on Democrats and Republicans

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

obyl

-0.72

icles

-0.69

iddles

-0.68

weather

-0.65

oken

-0.64

obar

-0.63

urious

-0.62

iffs

-0.62

ouston

-0.61

ysics

-0.61

POSITIVE LOGITS

 Party

0.64

maid

0.59

 standby

0.55

 stronghold

0.54

 operative

0.53

naissance

0.53

 counterpart

0.53

hood

0.53

å£«

0.52

geist

0.52

Activations Density 7.934%

mentions of political affiliations, specifically focusing on Democrats and Republicans

No Comments

No Known Activations