INDEX

Explanations

descriptions or mentions of personality traits

oai_token-act-pair · gpt-3.5-turbo

references to personality traits

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

YR

-0.78

nesday

-0.78

xon

-0.74

sonian

-0.73

heny

-0.73

thouse

-0.73

enegger

-0.72

WARE

-0.71

anus

-0.71

ichen

-0.70

POSITIVE LOGITS

 traits

1.38

 trait

1.28

 quirks

1.24

 disorder

1.17

 personality

0.96

 disorders

0.93

istics

0.91

istically

0.90

 charact

0.89

 personalities

0.89

Activations Density 0.035%

descriptions or mentions of personality traits

references to personality traits

No Comments

No Known Activations