INDEX

Explanations

model training and building

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ে

3.80

ೂರ್ವ

2.98

ревнова

2.79

к

2.74

ανο

2.66

sampler

2.62

screw

2.62

বিত্র

2.60

明け

2.59

sister

2.57

POSITIVE LOGITS

AndView

2.26

 نم

2.24

 tumb

2.23

ーション

2.23

argout

2.07

ี่ยม

2.05

argin

2.01

oso

2.00

erasan

2.00

spi

1.99

Activations Density 0.038%