INDEX

Explanations

model weights, script, model, supported, market, based

The neuron is detecting floating-point numeric tokens (i.e. numbers containing a decimal point).

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

0.81

 republik

0.74

 называ

0.74

 восем

0.73

 verification

0.71

stilling

0.71

verification

0.70

 названия

0.69

Reveal

0.68

tí

0.68

POSITIVE LOGITS

してた

0.93

 veines

0.89

 endroits

0.89

 ongles

0.86

 réunis

0.85

㈠

0.85

してる

0.84

 équipes

0.84

ق

0.84

isées

0.83

Activations Density 0.001%