INDEX

Explanations

the word "holiday" at different intensities

oai_token-act-pair · gpt-3.5-turbo

references to holidays

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

adle

-0.69

 Daniels

-0.69

 Aden

-0.68

nington

-0.68

ected

-0.67

efficients

-0.64

ioch

-0.64

anos

-0.63

anco

-0.63

ector

-0.62

POSITIVE LOGITS

 holidays

1.12

 holiday

1.03

idays

0.90

 Holiday

0.89

 weekend

0.88

iday

0.80

 decorations

0.78

 vacation

0.77

 vacations

0.75

 season

0.75

Activations Density 0.008%

the word "holiday" at different intensities

references to holidays

No Comments

No Known Activations