INDEX

Explanations

results and findings comparison

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 comienzos

-0.97

 erreur

-0.91

 inicios

-0.87

initView

-0.82

niaus

-0.82

larımız

-0.82

 suced

-0.82

 cím

-0.81

CMAKE

-0.81

 laget

-0.80

POSITIVE LOGITS

 then

2.17

 results

1.64

Then

1.48

然后

1.40

 потім

1.34

 затем

1.32

Results

1.31

then

1.30

 потом

1.29

 findings

1.28

Activations Density 0.144%