INDEX

Explanations

references or mentions of people within various contexts

oai_token-act-pair · gpt-3.5-turbo

mentions of "people"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 return

-0.72

 offense

-0.72

 finishing

-0.66

 salv

-0.66

 avail

-0.64

 turret

-0.64

 pinch

-0.63

 grade

-0.63

 starter

-0.63

 favor

-0.63

POSITIVE LOGITS

People

3.38

 People

2.36

people

2.27

 PEOPLE

1.82

Someone

1.70

Everyone

1.63

Everybody

1.60

Nobody

1.59

Kids

1.58

Women

1.53

Activations Density 0.018%

references or mentions of people within various contexts

mentions of "people"

No Comments

No Known Activations