INDEX

Explanations

commands or features related to a specific electronic device, specifically the Amazon Echo

oai_token-act-pair · gpt-3.5-turbo

mentions of Amazon Echo and Alexa products

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

src

-0.71

ARDS

-0.68

 Mehran

-0.67

arians

-0.67

eways

-0.67

 Wikimedia

-0.66

ammed

-0.66

aka

-0.66

atern

-0.65

ASC

-0.65

POSITIVE LOGITS

 Echo

1.06

 Chaser

0.79

 Nest

0.78

 mosqu

0.71

ertodd

0.71

actory

0.71

lore

0.71

piece

0.69

lys

0.68

 Lauder

0.68

Activations Density 0.007%

commands or features related to a specific electronic device, specifically the Amazon Echo

mentions of Amazon Echo and Alexa products

No Comments

No Known Activations