INDEX

Explanations

phrases related to advertisements

oai_token-act-pair · gpt-3.5-turbo

instances of advertisements

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

erva

-0.70

»Ĵ

-0.69

quartered

-0.68

ĪĴ

-0.66

stood

-0.66

borg

-0.64

folk

-0.63

cius

-0.63

roots

-0.62

ciplinary

-0.61

POSITIVE LOGITS

 Thumbnails

0.95

 Advertisement

0.79

VERTISEMENT

0.76

Interstitial

0.72

Skip

0.70

 advertisement

0.69

advertisement

0.69

ļéĨĴ

0.68

 toggle

0.66

 Transcript

0.65

Activations Density 0.008%

phrases related to advertisements

instances of advertisements

No Comments

No Known Activations