INDEX

Explanations

agent-based model

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

解决方案

-0.81

IKO

-0.76

eavour

-0.72

 Stavanger

-0.71

iembrie

-0.71

 Lazar

-0.71

dą

-0.70

 ischemic

-0.70

рики

-0.70

bino

-0.69

POSITIVE LOGITS

 agent

2.73

 Agent

2.31

Agent

2.20

agent

2.13

 agents

2.02

 Agents

1.70

 AGENT

1.66

Agents

1.66

 agente

1.65

 simulation

1.63

Activations Density 0.093%