INDEX

Explanations

phrases related to accusation or blame

oai_token-act-pair · gpt-3.5-turbo

statements and accusations related to criminal behavior or misconduct

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Wilmington

-0.83

aleigh

-0.74

 Jacksonville

-0.73

 Newport

-0.72

cone

-0.70

 Appalach

-0.68

 Boone

-0.67

 coasts

-0.67

 Warehouse

-0.66

 Midwest

-0.66

POSITIVE LOGITS

ÄŁ

1.33

ãĢİ

1.29

Ä

1.29

 ãĢĮ

1.24

 à¤

1.20

Å

1.20

 Kejriwal

1.20

 Allaah

1.18

Aad

1.18

ãĢĲ

1.15

Activations Density 0.979%

phrases related to accusation or blame

statements and accusations related to criminal behavior or misconduct

No Comments

No Known Activations