INDEX

Explanations

descriptions including family members

oai_token-act-pair · gpt-3.5-turbo

references to 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 @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

IUM

-0.82

Ĥİ

-0.79

Fair

-0.75

Ĥª

-0.72

AE

-0.71

GMT

-0.70

 Collider

-0.69

aly

-0.69

WAY

-0.67

Premium

-0.67

POSITIVE LOGITS

 sister

1.28

 siblings

1.23

 sisters

1.18

 girlfriends

1.18

 nephew

1.18

 heir

1.17

 daughter

1.16

 daughters

1.16

 grandmother

1.15

 girlfriend

1.14

Activations Density 0.153%

descriptions including family members

references to family members and their relationships

No Comments

No Known Activations