INDEX

Explanations

references to specific individuals, particularly in legal or judicial contexts

oai_token-act-pair · gpt-3.5-turbo

references to specific justices and legal opinions in a judicial context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

Ã§

-0.75

orama

-0.72

>>\

-0.72

 unsustainable

-0.72

ufact

-0.70

 indo

-0.70

 airst

-0.68

 Azerb

-0.67

 Mayhem

-0.66

 Pyramid

-0.66

POSITIVE LOGITS

 Scalia

1.39

 justices

1.15

 Gins

1.05

Jud

0.95

 dissenting

0.91

 Brennan

0.87

 Sonia

0.86

 Gorsuch

0.85

Justice

0.84

OTUS

0.83

Activations Density 0.106%

references to specific individuals, particularly in legal or judicial contexts

references to specific justices and legal opinions in a judicial context

No Comments

No Known Activations