INDEX

Explanations

mentions of sneak peeks, previews, or teasers for upcoming content

oai_token-act-pair · gpt-3.5-turbo

mentions of upcoming events or previews

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 equilibrium

-0.68

Abstract

-0.67

 manipulation

-0.65

 suicides

-0.64

Values

-0.64

 usefulness

-0.64

 discredited

-0.64

 euth

-0.64

 disputes

-0.63

 abusive

-0.62

POSITIVE LOGITS

 teaser

1.07

 preview

0.96

 premiere

0.96

Tickets

0.94

 premie

0.93

 trailer

0.92

 Trailer

0.92

 previews

0.91

 synopsis

0.89

PHOTOS

0.89

Activations Density 0.801%

mentions of sneak peeks, previews, or teasers for upcoming content

mentions of upcoming events or previews

No Comments

No Known Activations