INDEX

Explanations

model answer code

This neuron fires on tokens composed primarily of question-mark punctuation (e.g. “?”, “??”, etc.).

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 pretrained

0.70

ans

0.64

)`;

0.64

>`;

0.63

⨯

0.62

BleStatus

0.59

 سرم

0.59

,’’

0.58

)`,

0.58

>{{

0.57

POSITIVE LOGITS

????

1.52

1.47

????????

1.47

 ????

1.33

???

1.32

.?

1.30

?.

1.30

??

1.30

?.

1.29

?????

1.27

Activations Density 0.073%