INDEX

Explanations

journal names and fields

The neuron detects journal‐name abbreviations (e.g. "J Cell Mol Med," "J Med Chem," "J Opt Soc Am A," etc.) in reference citations.

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

 raiſ

-1.74

神

-1.70

的時間

-1.65

 fremden

-1.64

 auß

-1.59

 нашел

-1.59

 каждом

-1.58

 bestehenden

-1.52

癆

-1.50

 äußer

-1.49

POSITIVE LOGITS

for

1.80

forKey

1.66

</tfoot>

1.65

.}(

1.63

1.55

 seperti

1.55

<bos>

1.52

*}[

1.44

撰

1.41

We

1.40

Activations Density 0.005%