INDEX

Explanations

mentions of geographic locations, specifically areas labeled as "Downtown."

oai_token-act-pair · gpt-3.5-turbo

references to downtown areas

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

-0.85

udic

-0.81

UL

-0.80

ula

-0.76

ulate

-0.74

gs

-0.73

uth

-0.73

abo

-0.73

neum

-0.72

gger

-0.71

POSITIVE LOGITS

 redevelopment

1.07

 Downtown

1.02

owntown

1.00

 skyline

0.92

 revital

0.86

 Plaza

0.85

 downtown

0.82

 Improvement

0.78

 storefront

0.75

 skysc

0.74

Activations Density 0.008%

mentions of geographic locations, specifically areas labeled as "Downtown."

references to downtown areas

No Comments

No Known Activations