INDEX

Explanations

base code prefixes

The neuron activates on occurrences of the token “base” (in code, config, or comments).

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

拋

-0.76

當中

-0.73

 surrounded

-0.73

 namentlich

-0.73

 Subject

-0.72

ådet

-0.72

 surround

-0.72

 ANOVA

-0.71

 bounded

-0.69

 likely

-0.69

POSITIVE LOGITS

 Base

1.24

getBase

1.23

 BASE

1.20

base

1.20

 getBase

1.10

Base

1.08

 base

1.06

BASE

0.96

 classes

0.94

 châssis

0.94

Activations Density 0.016%