INDEX

Explanations

phrases related to the origin or source of something

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 sanity

-0.63

visors

-0.63

²

-0.60

 attrition

-0.59

 timely

-0.58

 commons

-0.57

tsy

-0.56

 sorely

-0.55

azer

-0.54

 loos

-0.53

POSITIVE LOGITS

 beforehand

0.73

 Anyway

0.70

etc

0.68

abouts

0.67

<[

0.66

?????

0.65

ById

0.65

OST

0.64

igraph

0.64

iac

0.62

Activations Density 0.204%

phrases related to the origin or source of something

No Comments

No Known Activations