INDEX

Explanations

instructions requiring action

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

tir

1.94

iv

1.93

 orta

1.88

 télécharg

1.85

พลาด

1.80

দ্রোহ

1.79

 cose

1.78

ती

1.77

铃

1.76

 weniger

1.73

POSITIVE LOGITS

3.10

нном

3.01

eaa

2.90

eber

2.89

ン

2.67

2.62

dre

2.61

нным

2.57

دخل

2.53

sst

2.52

Activations Density 0.196%