INDEX

Explanations

mentions of the Pope

oai_token-act-pair · gpt-3.5-turbo

references to the Pope

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

-0.87

nesota

-0.86

yrinth

-0.82

awks

-0.74

rg

-0.73

é¾

-0.73

ahime

-0.72

mental

-0.71

erry

-0.71

ership

-0.70

POSITIVE LOGITS

 Francis

1.14

 Pope

0.95

Pope

0.88

 Benedict

0.86

 pont

0.83

 pope

0.78

 Father

0.75

orio

0.73

angelo

0.72

 Pablo

0.72

Activations Density 0.007%

mentions of the Pope

references to the Pope

No Comments

No Known Activations