INDEX

Explanations

mentions of specific popes

oai_token-act-pair · gpt-3.5-turbo

references to the Pope and related figures

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

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

ript

-0.78

nesota

-0.78

nels

-0.75

 Wonderland

-0.73

schild

-0.73

uld

-0.72

awks

-0.72

hooting

-0.72

ÑĮ

-0.71

Haw

-0.71

POSITIVE LOGITS

 Francis

1.27

 Benedict

1.10

 pope

0.92

Pope

0.90

XVI

0.89

 Vatican

0.87

 Pope

0.86

 Patriarch

0.85

 Archbishop

0.84

Franc

0.81

Activations Density 0.031%

mentions of specific popes

references to the Pope and related figures

No Comments

No Known Activations