INDEX

Explanations

movie titles featuring superheroes or popular movie franchises

oai_token-act-pair · gpt-3.5-turbo

references to superhero movies and franchise titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

lying

-0.83

 Adin

-0.76

Geh

-0.73

 Citiz

-0.69

 Pear

-0.67

uga

-0.65

isch

-0.64

itute

-0.64

rique

-0.63

seless

-0.63

POSITIVE LOGITS

 sequel

1.32

 sequels

1.29

 trailer

1.18

 Trailer

1.14

 trilogy

1.14

 films

1.13

 reboot

1.12

 franchise

1.12

 movie

1.11

 actor

1.11

Activations Density 0.202%

movie titles featuring superheroes or popular movie franchises

references to superhero movies and franchise titles

No Comments

No Known Activations