INDEX

Explanations

package name or copyright

The neuron detects the “@package” annotation (and related doc-block package/subpackage tags) in source‐code comments.

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

訁

-2.00

 olhando

-1.89

 ПОД

-1.86

 skvě

-1.84

いかがでしたか

-1.79

を中心

-1.75

 glittering

-1.72

 nová

-1.71

两天

-1.66

 nejlepší

-1.66

POSITIVE LOGITS

“

2.38

locene

2.03

1.74

 jari

1.70

1.66

ętr

1.64

quelize

1.63

幛

1.61

 Owing

1.59

Activations Density 0.006%