INDEX

Explanations

mentions of job titles and professional backgrounds

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

 suscept

-0.68

 catast

-0.68

 unbeliev

-0.66

 ceilings

-0.63

 whim

-0.63

 ultras

-0.63

asions

-0.60

 chaotic

-0.60

 Britons

-0.59

 chests

-0.59

POSITIVE LOGITS

 overseeing

0.81

 Assist

0.73

ective

0.73

/#

0.68

RECT

0.67

etter

0.65

iate

0.65

baugh

0.65

bernatorial

0.65

orney

0.65

Activations Density 12.501%

mentions of job titles and professional backgrounds

No Comments

No Known Activations