INDEX

Explanations

model / models

The neuron detects mentions of conceptual “model” (and its plural “models”) in the text.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

pulseira

-1.87

 henvis

-1.85

 alcanzó

-1.77



-1.70



-1.67

 cayó

-1.66

܃

-1.64

estima

-1.64

 forteller

-1.61

pendentif

-1.60

POSITIVE LOGITS

of

2.09

2.05

It

1.91

Your

1.86

as

1.77

You

1.72

.”

1.69

Our

1.67

られた

1.66

–

1.64

Activations Density 0.039%