INDEX

Explanations

instill, ingrained, into

This neuron detects verbs (and related forms) that mean “to instill or imbue” (i.e. impart a quality, value, or trait).

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

黢

-2.45

𓄹

-2.27

⽌

-2.17

罍

-2.14

鼯

-2.14

𓋼

-2.14

⢪

-2.13

瓩

-2.11

目錄

-2.05

 valentín

-2.03

POSITIVE LOGITS

selbe

2.00

 Welche

1.91

咐

1.84

issory

1.80

 provavelmente

1.73

Ndr

1.69

 quizá

1.67

即便是

1.64

在下

1.60

 talvez

1.59

Activations Density 0.005%