INDEX

Explanations

data structures like map, dictionary, hashmap

The neuron strongly activates on assignment operations (the “=”) in code.

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

 listOf

-1.10

listOf

-0.96

 list

-0.94

list

-0.94

метры

-0.90

cretion

-0.84

 Galapagos

-0.82

REACT

-0.81

setList

-0.78

[{

-0.76

POSITIVE LOGITS

{[

1.29

Map

1.09

new

1.08

{[

1.05

HashMap

1.05

 HashMap

1.04

Hashtable

0.96

{};

0.95

 Dictionary

0.87

 efficiencies

0.85

Activations Density 0.015%