INDEX

Explanations

words related to specific locations and neighborhoods within San Francisco

oai_token-act-pair · gpt-3.5-turbo

elements related to crime incidents and law enforcement in San Francisco

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 Maced

-0.98

 Ukip

-0.93

 Macedonia

-0.89

 Ukrainians

-0.89

 Eston

-0.88

 Estonia

-0.86

Äį

-0.84

 Belarus

-0.84

 Seym

-0.83

 Thib

-0.83

POSITIVE LOGITS

 Oakland

2.33

Oak

2.02

 BART

1.93

San

1.78

 Berkeley

1.78

ameda

1.71

SF

1.68

Bay

1.64

San

1.60

 Hayward

1.58

Activations Density 0.316%

words related to specific locations and neighborhoods within San Francisco

elements related to crime incidents and law enforcement in San Francisco

No Comments

No Known Activations