INDEX

Explanations

mentions of an organization or event called "Indy."

oai_token-act-pair · gpt-3.5-turbo

references to the Indy Car series and related events

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

 Gleaming

-0.83

istics

-0.77

atin

-0.73

 Imran

-0.72

orative

-0.71

aunder

-0.71

utenberg

-0.70

utations

-0.69

atche

-0.66

interstitial

-0.66

POSITIVE LOGITS

 Eleven

1.09

 Lights

0.89

 Indy

0.82

Car

0.80

ML

0.79

gee

0.75

tti

0.74

 Apost

0.71

Ind

0.71

Star

0.70

Activations Density 0.013%

mentions of an organization or event called "Indy."

references to the Indy Car series and related events

No Comments

No Known Activations