INDEX

Explanations

family relationships

oai_token-act-pair · gpt-3.5-turbo

phrases that mention family members and their relationships

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Applications

-0.76

ieties

-0.74

ensus

-0.72

places

-0.68

ģ«

-0.65

AE

-0.65

oud

-0.65

Times

-0.64

Vers

-0.64

Words

-0.64

POSITIVE LOGITS

 daughter

1.25

 niece

1.24

 granddaughter

1.21

 mistress

1.18

 grandson

1.17

 girlfriend

1.16

 wife

1.09

 sister

1.08

 heir

1.06

 cousin

1.06

Activations Density 0.159%

family relationships

phrases that mention family members and their relationships

No Comments

No Known Activations