INDEX

Explanations

great for

This neuron detects strong positive evaluative adjectives (like “great,” “perfect,” “suitable”) that praise or endorse something as ideal or highly recommended.

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

-1.82

 kön

-1.70

it

-1.63

锨

-1.61

neling

-1.61

 coreana

-1.61

 augus

-1.57

埏

-1.55

its

-1.54

 rajut

-1.51

POSITIVE LOGITS

for

2.23

to

1.56

↵↵

1.31

渙

1.29

同样

1.27

 типов

1.27

最初の

1.27

 folgendes

1.22

这

1.21

1.20

Activations Density 0.030%