INDEX

Explanations

mentions of people's awareness or knowledge of specific information or situations

oai_token-act-pair · gpt-3.5-turbo

instances of the word "aware" indicating recognition or consciousness of information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

Textures

-0.76

mini

-0.62

ueller

-0.62

artney

-0.61

irrel

-0.61

MER

-0.60

nic

-0.59

mer

-0.59

cell

-0.59

PET

-0.58

POSITIVE LOGITS

ledged

1.17

ledge

0.95

 aware

0.90

ness

0.86

 Aware

0.84

isance

0.84

 Lauder

0.81

eering

0.78

itaire

0.77

icip

0.76

Activations Density 0.019%

mentions of people's awareness or knowledge of specific information or situations

instances of the word "aware" indicating recognition or consciousness of information

No Comments

No Known Activations