INDEX

Explanations

organizations and legal terms related to human rights

oai_token-act-pair · gpt-3.5-turbo

mentions of human rights organizations and related advocacy

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

jun

-0.67

 selves

-0.67

cyl

-0.66

classes

-0.66

crew

-0.66

 Styles

-0.65

brew

-0.63

realDonaldTrump

-0.62

berry

-0.61

steam

-0.61

POSITIVE LOGITS

 GOODMAN

0.89

NGO

0.85

 Amnesty

0.83

 activists

0.80

 UNHCR

0.78

SOS

0.77

 Farrell

0.76

ighed

0.75

 activist

0.73

 Equality

0.71

Activations Density 0.142%

organizations and legal terms related to human rights

mentions of human rights organizations and related advocacy

No Comments

No Known Activations