INDEX

Explanations

hydrophilic and electrophilic suffixes

The neuron detects scientific terms ending with "-philic" or "-phile", particularly related to chemical and biological properties.

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

！”

-2.28

-2.09

 continúas

-2.09

平凡

-2.02

 finnas

-1.98

 важно

-1.93

KELEY

-1.89

幼稚

-1.88

 включая

-1.88

 imma

-1.84

POSITIVE LOGITS

んびり

2.41

</h1>

2.27

</sub>

2.19

떳

2.08

 tetapi

2.08

</h3>

2.08

";

2.05

雱

2.05

 spirited

2.02

})

2.00

Activations Density 0.023%