INDEX

Explanations

json object parsing

The neuron activates on mentions of JSON‐related class or type names (e.g. JSONObject, JSONArray, JsonObject, JsonNode).

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

 with

-1.41

on

-1.38

to

-1.29

 begin

-1.21

 from

-1.20

for

-1.18

 these

-1.18

 have

-1.15

 each

-1.14

 even

-1.13

POSITIVE LOGITS

ー

1.06

abilité

1.05

 abel

1.05

TRIB

1.05

 líderes

1.05

疖

1.05

 appeler

1.05

 appelés

1.04

 joyería

1.03

 mejoras

1.02

Activations Density 0.004%