INDEX

Explanations

phrases related to movies, shows, or adaptations based on existing works

oai_token-act-pair · gpt-3.5-turbo

references to films and adaptations based on stories and other source material

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

wcs

-0.83

ciating

-0.78

twitch

-0.77

isphere

-0.72

gpu

-0.72

OTAL

-0.72

erity

-0.71

Interstitial

-0.70

hift

-0.70

FML

-0.70

POSITIVE LOGITS

 novels

1.37

 bestselling

1.20

 memoir

1.18

 novel

1.16

 autobiography

1.10

 autobi

1.07

 screenplay

1.04

 manga

1.03

 novelist

1.01

 Pulitzer

0.93

Activations Density 0.220%

phrases related to movies, shows, or adaptations based on existing works

references to films and adaptations based on stories and other source material

No Comments

No Known Activations