INDEX

Explanations

specific mentions of metropolitan areas such as Seattle, Houston, and Los Angeles

oai_token-act-pair · gpt-3.5-turbo

references to specific geographic areas

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

Not in Any Lists

No Comments

Negative Logits

artz

-0.81

ificial

-0.80

FILE

-0.77

forestation

-0.72

 PROGRAM

-0.71

estamp

-0.71

icial

-0.70

Files

-0.70

ongyang

-0.67

skill

-0.67

POSITIVE LOGITS

 suburbs

0.87

 Borough

0.84

 Parkway

0.82

 area

0.82

 northeast

0.79

 median

0.78

 southwest

0.78

 northwest

0.78

 Northeast

0.76

 west

0.75

Activations Density 0.043%

specific mentions of metropolitan areas such as Seattle, Houston, and Los Angeles

references to specific geographic areas

No Comments

No Known Activations