INDEX

Explanations

references to literature

oai_token-act-pair · gpt-3.5-turbo

references to various forms of literature

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

fty

-0.64

addafi

-0.64

 twitch

-0.63

 Maurit

-0.62

ebus

-0.61

por

-0.60

ermanent

-0.59

heed

-0.57

adjust

-0.57

Gry

-0.56

POSITIVE LOGITS

 literature

0.88

emis

0.80

istry

0.79

DragonMagazine

0.75

geist

0.74

 RELE

0.72

 writ

0.71

cox

0.70

uggest

0.69

wallet

0.68

Activations Density 0.021%

references to literature

references to various forms of literature

No Comments

No Known Activations