INDEX

Explanations

types of soups

oai_token-act-pair · gpt-3.5-turbo

references to various types of soup

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

uate

-0.68

inel

-0.67

rium

-0.65

rics

-0.62

 Surveillance

-0.62

 Hubble

-0.62

 Democr

-0.62

 Planetary

-0.60

 WATCHED

-0.60

ussen

-0.60

POSITIVE LOGITS

 soup

1.29

 Soup

1.18

bowl

1.11

 kitchens

0.98

 nood

0.86

Ã©e

0.84

 broth

0.84

 bowls

0.84

 kitchen

0.82

Ã§

0.81

Activations Density 0.010%

types of soups

references to various types of soup

No Known Activations