INDEX

Explanations

words related to high sales or popularity

oai_token-act-pair · gpt-3.5-turbo

terms related to sales and popularity of products, particularly in the context of bestsellers

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

sterdam

-0.83

laus

-0.72

hra

-0.70

 sshd

-0.69

 Ping

-0.66

involved

-0.66

anky

-0.65

 Daylight

-0.65

endez

-0.64

 Security

-0.63

POSITIVE LOGITS

selling

0.95

seller

0.88

atform

0.84

sell

0.84

Interstitial

0.83

 publications

0.82

 receipts

0.81

 bestselling

0.80

 books

0.78

 endorsements

0.77

Activations Density 0.030%

words related to high sales or popularity

terms related to sales and popularity of products, particularly in the context of bestsellers

No Comments

No Known Activations