You could use the ‘commit’ id to join with GitHub Archive.

SELECT COUNT(*) commiters_sample
FROM (
  SELECT actor.login
  FROM [bigquery-public-data:github_repos.sample_commits] a
  JOIN (
    SELECT JSON_EXTRACT_SCALAR(payload, '$.commits[0].sha') sha, actor.login
    FROM [githubarchive:day.20160101] b
    WHERE type='PushEvent'
  ) b
  ON a.commit=b.sha
  GROUP BY 1
)
LIMIT 10

Undoubtedly it could be easier — but at least his works :).