INDEX

Explanations

mentions of when something was first reported or created

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 Canaver

-0.66

 Gould

-0.64

sav

-0.62

SOM

-0.62

ourge

-0.61

urgy

-0.60

antics

-0.59

 Doomsday

-0.58

vor

-0.57

âĹ¼

-0.56

POSITIVE LOGITS

 baseman

1.22

 responders

1.18

 blush

0.90

 glance

0.89

ly

0.75

arthed

0.70

 lady

0.70

ially

0.69

ebin

0.68

 impressions

0.68

Activations Density 0.055%

mentions of when something was first reported or created

No Comments

No Known Activations