Open Models Almanac
Methodsdata curation

specific method · filed under data curation

Deduplication and filtering

A curation method that removes duplicate, junk, or low-quality data and may support safety objectives.

The sources on this page only mention it in passing. The line above is the extraction's one-line gloss. See data curation for the methods filed beside it.

Also called Training data deduplication and filtering.

sources
2
model
1
lab adopt it
1
strongest
used

Documented in

Thinking Machines Lab

Evidence

2 quotes from 2 sources, one card per model, each opening with its strongest.

How sources treat it
used 2

Inkling

Thinking Machines Labadopted2 sources

may include deduplication and filtering to remove junk or other low-quality data, or to advance safety or other objectives

useddata curationin InklingThinking Machines Lab
1 more from this model

may include deduplication and filtering to remove junk or other low-quality data, or to advance safety or other objectives

useddata curationin InklingThinking Machines Lab
↑