INDEX

Explanations

references to the action of taking selfies

oai_token-act-pair · gpt-3.5-turbo

mentions of selfies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

quartered

-0.81

ajor

-0.80

ifted

-0.77

assian

-0.73

orf

-0.72

rawdownloadcloneembedreportprint

-0.72

mental

-0.70

Ļ½

-0.68

iant

-0.68

profits

-0.67

POSITIVE LOGITS

 selfies

1.19

 selfie

1.12

 pics

0.90

 Doodle

0.83

 Bieber

0.74

 booth

0.74

 PHOTO

0.73

 Snapchat

0.69

photos

0.68

cam

0.68

Activations Density 0.014%

references to the action of taking selfies

mentions of selfies

No Comments

No Known Activations