INDEX

Explanations

profession-related words such as "builder," "photographer," and "teacher."

oai_token-act-pair · gpt-3.5-turbo

various types of professionals and their attributes or roles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

edIn

-0.72

 Matter

-0.71

 domains

-0.68

ouls

-0.67

uld

-0.66

Vote

-0.66

 Hath

-0.64

 Horses

-0.64

laws

-0.64

 Trails

-0.62

POSITIVE LOGITS

 extraord

1.42

digy

1.11

 specializing

0.96

 himself

0.93

 whose

0.87

 herself

0.86

izer

0.85

 whis

0.83

who

0.81

ess

0.80

Activations Density 0.222%

profession-related words such as "builder," "photographer," and "teacher."

various types of professionals and their attributes or roles

No Comments

No Known Activations