INDEX

Explanations

instances of the word "juice," specifically focusing on its presence in different contexts within the document

oai_token-act-pair · gpt-3.5-turbo

references to various types of juice

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

sighted

-0.75

 WATCHED

-0.72

igmat

-0.69

humane

-0.65

ear

-0.64

alty

-0.63

yrinth

-0.63

atern

-0.62

gotten

-0.62

ndum

-0.62

POSITIVE LOGITS

 juice

1.35

 juices

1.20

 Juice

1.19

wine

0.93

fruit

0.89

xtap

0.89

 Beans

0.83

tip

0.79

 oranges

0.78

usha

0.77

Activations Density 0.009%

instances of the word "juice," specifically focusing on its presence in different contexts within the document

references to various types of juice

No Comments

No Known Activations