INDEX

Explanations

code variable assignments

This neuron detects comma‐separated pairs of loop variables (tuple unpacking) in for‐loops or comprehensions, e.g. “for k, v in …” or “for idx, Item in …”.

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

irée

-1.65



-1.57

猡

-1.55



-1.53

 volvió

-1.53

了几

-1.52

 conhecida

-1.52

-1.51

更何况

-1.48

 diámetro

-1.48

POSITIVE LOGITS

of

2.33

That

2.05

It

1.98

Then

1.88

Clearly

1.81

Also

1.80

These

1.80

This

1.76

All

1.73

Instead

1.73

Activations Density 0.005%