INDEX

Explanations

titles of TV shows and movies

oai_token-act-pair · gpt-3.5-turbo

references to popular culture characters and titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 mosqu

-0.75

 conflic

-0.72

 warr

-0.68

antha

-0.66

ometimes

-0.65

 pestic

-0.64

 obser

-0.64

ĪĴ

-0.63

perty

-0.63

MIS

-0.63

POSITIVE LOGITS

 Awakens

0.93

 Reloaded

0.83

 trilogy

0.79

 soundtrack

0.73

 Trilogy

0.73

idge

0.73

 Unleashed

0.71

âĦ¢:

0.69

 Collection

0.68

 Metatron

0.67

Activations Density 0.142%

titles of TV shows and movies

references to popular culture characters and titles

No Comments

No Known Activations