INDEX

Explanations

This neuron activates specifically on floating-point/decimal number tokens.

The neuron detects strongly evaluative or emphatic words (intensifying adjectives/adverbs and sentiment-laden descriptors).

New Auto-Interp

Configuration

Prompts (Dashboard)

238,145 prompts, 512 tokens each

Dataset (Dashboard)

lmsys + oasst1

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

ly

0.38

な

0.34

fully

0.33

amente

0.32

ity

0.31

 এসব

0.31

 усі

0.31

scapes

0.30

<unused2182>

0.30

 ترین

0.30

POSITIVE LOGITS

 Вы

0.39

 Ayrıca

0.34

 Emulator

0.30

</b>

0.30

 menino

0.30

 chromatographed

0.30

는

0.29

 Sabtu

0.29

 sogenannten

0.28

 wrecked

0.28

Activations Density 0.000%