INDEX

Explanations

mentions of specific locations, specifically related to Beverly Hills

oai_token-act-pair · gpt-3.5-turbo

references to locations or proper nouns related to Beverly Hills

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

*/(

-1.07

ebook

-0.98

udic

-0.79

emic

-0.77

idem

-0.76

OPLE

-0.68

ems

-0.68

ittens

-0.67

urgy

-0.65

drawn

-0.65

POSITIVE LOGITS

 Hills

1.14

 Beverly

1.07

 Crusher

0.98

 Boulevard

0.87

ãĤ©

0.84

 Lumpur

0.79

 Blvd

0.78

 Dame

0.76

 Hilton

0.75

 Lauder

0.73

Activations Density 0.009%

mentions of specific locations, specifically related to Beverly Hills

references to locations or proper nouns related to Beverly Hills

No Comments

No Known Activations