INDEX

Explanations

numbers and codes

The neuron responds to alphanumeric technical identifiers (e.g. gene or strain codes, model numbers, or version strings) that combine letters and digits.

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

but

-2.05

in

-1.99

~~~~~~~

-1.93

👎

-1.87

if

-1.87

🤣🤣

-1.82

😹

-1.81

😫

-1.80

~~~~~

-1.77

🤢

-1.77

POSITIVE LOGITS

</b>

2.41

 your

2.08

 their

1.95

1.88

 malheure

1.73

 whats

1.71

CHREIB

1.66

晅

1.61

我只是

1.59

.$_

1.58

Activations Density 0.078%