INDEX

Explanations

instances of the word "authority."

oai_token-act-pair · gpt-3.5-turbo

references to the concept of authority

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

 Lovely

-0.78

 Resist

-0.76

 Shal

-0.75

 Leica

-0.71

Lens

-0.68

Ale

-0.67

 Taste

-0.65

 Indust

-0.63

 Schwar

-0.63

 Alto

-0.63

POSITIVE LOGITS

 authority

1.26

Reviewer

1.04

 delegated

0.93

 vested

0.89

 conferred

0.86

 nomine

0.85

authorized

0.83

 confir

0.82

ignty

0.80

 authorities

0.78

Activations Density 0.010%

instances of the word "authority."

references to the concept of authority

No Comments

No Known Activations