INDEX

Explanations

locations mentioned in the state of California, particularly those including "San" in the name

oai_token-act-pair · gpt-3.5-turbo

mentions of San Francisco and other related San locations

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

ilater

-0.89

ï¸ı

-0.74

dash

-0.69

xual

-0.68

anwhile

-0.66

ICAL

-0.66

DonaldTrump

-0.65

 clipboard

-0.65

 tremend

-0.64

Reloaded

-0.64

POSITIVE LOGITS

San

1.31

 Francisco

1.24

 Diego

1.11

San

1.10

 Antonio

0.98

ctuary

0.92

 Angeles

0.90

eteenth

0.89

ction

0.87

ushi

0.86

Activations Density 0.014%

locations mentioned in the state of California, particularly those including "San" in the name

mentions of San Francisco and other related San locations

No Comments

No Known Activations