INDEX

Explanations

mentions of actors and related terms

oai_token-act-pair · gpt-3.5-turbo

instances of the word 'actor' and related roles in the context of performance and entertainment

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

etheless

-0.73

£ı

-0.72

izoph

-0.68

wart

-0.66

fty

-0.66

rontal

-0.64

Ã§a

-0.63

iness

-0.61

oys

-0.61

ppings

-0.60

POSITIVE LOGITS

 portraying

1.22

 audition

0.91

 portray

0.88

ãĤ´

0.83

 actor

0.82

 Leonardo

0.81

 portrays

0.81

 Judd

0.81

 rehears

0.79

 starring

0.79

Activations Density 0.058%

mentions of actors and related terms

instances of the word 'actor' and related roles in the context of performance and entertainment

No Comments

No Known Activations