INDEX

Explanations

phrases related to expressing beliefs, warnings, and predictions

oai_token-act-pair · gpt-3.5-turbo

declarative statements expressing opinions or beliefs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

DragonMagazine

-0.92

orah

-0.85

operator

-0.83

him

-0.78

yssey

-0.76

 Investigator

-0.76

nesday

-0.75

ractor

-0.75

mmmm

-0.73

arah

-0.72

POSITIVE LOGITS

 governments

1.40

 nations

1.30

 democracies

1.28

 Britons

1.20

 countries

1.20

 societies

1.19

 Americans

1.17

 economies

1.15

 policymakers

1.14

 Governments

1.14

Activations Density 0.503%

phrases related to expressing beliefs, warnings, and predictions

declarative statements expressing opinions or beliefs

No Comments

No Known Activations