INDEX

Explanations

metabolites produced

The neuron identifies the word "metabolites" in scientific or medical text.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

等等

-2.67

絰

-2.44

 Beſ

-2.44

鲠

-2.31

あるので

-2.25

 Biblia

-2.22

ParallelGroup

-2.16

 vous

-2.16

 silueta

-2.16

-2.14

POSITIVE LOGITS

秝

2.88

section

2.61

ブログ村

2.56

余談

2.45

珢

2.30

2.28

2.22

屁股

2.19

 Они

2.16

Activations Density 0.009%