INDEX

Explanations

sentence starts after punctuation

The neuron chiefly detects standalone numeric tokens (e.g. years, counts, or other numeral values).

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

 ſie

-1.19

=]

-1.08

 этом

-1.06

⿻

-1.06

 bluz

-1.05

🙏🙏

-1.04

_'.$

-1.02

 then

-1.02

-1.01

 kazak

-0.99

POSITIVE LOGITS

<eos>

1.36

six

1.09

Nevertheless

1.07

 with

1.06

 There

1.05

剁

1.02

 standard

1.02

 severe

1.00

他想

1.00

 specific

0.98

Activations Density 0.015%