INDEX

Explanations

video titles urging viewers to watch

oai_token-act-pair · gpt-3.5-turbo

content that promotes or emphasizes the act of watching certain videos or shows

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

 recl

-0.68

 academia

-0.67

 nowhere

-0.66

urally

-0.66

clusively

-0.65

 plagiar

-0.65

 humble

-0.64

 beet

-0.63

 sucker

-0.63

 disinteg

-0.63

POSITIVE LOGITS

 WATCH

1.41

 WATCHED

1.08

 Thumbnails

1.06

 Watching

1.06

 VIDEOS

0.99

WATCH

0.94

]}

0.92

 Watch

0.89

LIST

0.81

IELD

0.81

Activations Density 0.005%

video titles urging viewers to watch

content that promotes or emphasizes the act of watching certain videos or shows

No Comments

No Known Activations