INDEX

Explanations

mentions of movie or video game franchises

oai_token-act-pair · gpt-3.5-turbo

references to popular media franchises and their related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

egg

-0.74

student

-0.72

Aid

-0.71

partisan

-0.71

ĵĺ

-0.70

trust

-0.69

¬¼

-0.69

agogue

-0.69

cipled

-0.68

trump

-0.68

POSITIVE LOGITS

 trilogy

1.03

 sequels

1.01

 reboot

0.96

 canon

0.93

 remake

0.92

 sequel

0.92

 Capcom

0.91

 adaptation

0.90

 cinematic

0.89

 manga

0.87

Activations Density 0.216%

mentions of movie or video game franchises

references to popular media franchises and their related terms

No Comments

No Known Activations