INDEX

Explanations

instances of words related to legal authority and power

oai_token-act-pair · gpt-3.5-turbo

references to governmental powers and authorities

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 Sakuya

-0.68

adish

-0.67

 Lect

-0.66

Rum

-0.65

 Remastered

-0.65

Lens

-0.64

hiba

-0.64

Rum

-0.64

Gil

-0.64

 Stain

-0.64

POSITIVE LOGITS

 delegated

1.16

 vested

1.02

 granted

0.99

 conferred

0.96

 levers

0.94

 discretionary

0.92

 authority

0.91

 discretion

0.89

 empowered

0.84

 powers

0.83

Activations Density 0.081%

instances of words related to legal authority and power

references to governmental powers and authorities

No Known Activations