INDEX

Explanations

mentions of privilege, especially in social contexts

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

tra

-0.83

ood

-0.77

agh

-0.74

ICA

-0.74

thumbnails

-0.71

urgy

-0.70

rum

-0.69

enegger

-0.68

NCT

-0.67

hani

-0.66

POSITIVE LOGITS

ilege

1.39

ileged

1.21

 privilege

1.11

 Priv

1.01

 privileges

0.85

 holders

0.84

 privileged

0.79

Priv

0.77

SPONSORED

0.76

 afforded

0.75

Activations Density 0.062%

mentions of privilege, especially in social contexts

No Comments

No Known Activations