INDEX

Explanations

mentions of government officials and political positions

oai_token-act-pair · gpt-3.5-turbo

references to government officials or cabinet members

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

owicz

-0.71

Gim

-0.68

Ô

-0.67

 GOODMAN

-0.66

ibaba

-0.65

thia

-0.63

Ïģ

-0.63

iesta

-0.62

amples

-0.61

 Wear

-0.61

POSITIVE LOGITS

 secretaries

1.23

 ministers

1.11

 appointments

1.03

chairs

0.97

 chairs

0.93

 nominees

0.91

 minister

0.90

 shuffle

0.90

 Ministers

0.89

 members

0.87

Activations Density 0.021%

mentions of government officials and political positions

references to government officials or cabinet members

No Comments

No Known Activations