INDEX

Explanations

predicting future outcomes

predicting outcomes or words

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

ра

0.61

aş

0.53

БУ

0.51

es

0.50

ia

0.50

ürdig

0.49

 существуют

0.48

Có

0.47

evident

0.47

ла

0.47

POSITIVE LOGITS

 predict

1.19

 predicts

1.10

 예측

1.08

 prediction

1.04

predict

1.01

Predict

1.01

 भविष्यवाणी

1.00

 predicting

0.99

预测

0.99

予測

0.95

Activations Density 0.053%