train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v2/split_1043-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json

{
  "num_tokens": 38901586,
  "num_truncated_tokens": 38870833
}