INDEX

Explanations

references to the TV series "Game of Thrones" and its related characters

oai_token-act-pair · gpt-3.5-turbo

references to the "Game of Thrones" series and its characters

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

trak

-0.93

berman

-0.72

ntil

-0.72

aneous

-0.71

xon

-0.70

ees

-0.70

igating

-0.69

iating

-0.68

aneously

-0.66

esville

-0.66

POSITIVE LOGITS

 Thrones

1.36

 Lann

1.14

 Reborn

0.79

 Targ

0.79

 compr

0.78

 fandom

0.73

 premiered

0.72

 Dracula

0.70

HBO

0.69

ullivan

0.69

Activations Density 0.041%

references to the TV series "Game of Thrones" and its related characters

references to the "Game of Thrones" series and its characters

No Comments

No Known Activations