INDEX

Explanations

phrases related to power or control

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate governance or control by various entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

naissance

-0.92

apest

-0.83

ased

-0.82

asy

-0.81

enery

-0.81

icter

-0.78

raq

-0.78

enth

-0.78

bia

-0.78

assment

-0.76

POSITIVE LOGITS

 virtue

0.99

laws

0.81

 fiat

0.76

 sheer

0.74

products

0.74

 statute

0.73

 various

0.71

 varying

0.71

 multiplying

0.69

 gravity

0.67

Activations Density 0.109%

phrases related to power or control

phrases that indicate governance or control by various entities

No Comments

No Known Activations