Issues

• get_aggregated_results(): per-category, overall and aggregation_json averages are computed over whichever tasks ran, while Core_{version}/Extended_{version} report N/A for the same missing tasks

  mlfoundations/dclm#121 · Filter by author shaurya416 opened on Sep 25, 2026

• Problems when replicating the 400M-1x experiment

  mlfoundations/dclm#120 · Filter by author JerryYan123 opened on Mar 19, 2026

• DCLM-refinedweb not available: denied access aws s3 bucket

  mlfoundations/dclm#119 · Filter by author AROURI31 opened on Mar 8, 2026

• Running on test json.gz file

  mlfoundations/dclm#118 · Filter by author Aditya-Manjunatha opened on Feb 3, 2026

• Wrong random baseline value for boolq

  mlfoundations/dclm#117 · Filter by author fcmomentum opened on Dec 2, 2025

• CORE centered accuracy calculation and random baseline

  mlfoundations/dclm#114 · Filter by author karpathy opened on Jul 22, 2025
  2

• 3B-1x baseline results

  mlfoundations/dclm#112 · Filter by author yuzc19 opened on Mar 25, 2025

• 400M Model Weights

  mlfoundations/dclm#105 · Filter by author ethanlshen opened on Jan 14, 2025

• Dataset for training 1B-5x model in Table 1

  mlfoundations/dclm#104 · Filter by author LeoXinhaoLee opened on Dec 29, 2024

• denied access while copying shards from aws s3 bucket

  mlfoundations/dclm#103 · Filter by author emirkaan5 opened on Dec 13, 2024
  3

• Searching DCLM-baseline

  mlfoundations/dclm#101 · Filter by author chtmp223 opened on Dec 5, 2024

• Build my own model and use DCLM-1B training script and dataset.

  mlfoundations/dclm#95 · Filter by author windbar778 opened on Nov 17, 2024

• tokenization memory usage

  mlfoundations/dclm#88 · Filter by author brian-ham opened on Oct 13, 2024
  1

• Availability of DCLM data mixes used in figure 3

  mlfoundations/dclm#77 · Filter by author IanMagnusson opened on Sep 13, 2024
  4

  

• The dataset for training fastText OH-2.5 +ELI5 text classifier

  mlfoundations/dclm#75 · Filter by author yqy2001 opened on Sep 10, 2024
  5

  

• Training data of model-based filtering

  mlfoundations/dclm#74 · Filter by author Yu-Shi opened on Sep 6, 2024
  3

  

• deduplication removes 98% of my data

  mlfoundations/dclm#71 · Filter by author Yu-Shi opened on Sep 2, 2024
  2

  

• What is the pretrain scripts?

  mlfoundations/dclm#68 · Filter by author mathfinder opened on Aug 30, 2024
  12

  

• TypeError: Couldn't cast array of type

  mlfoundations/dclm#66 · Filter by author shizhediao opened on Aug 28, 2024
  2

  

• Any plans to release pools after refinedweb heuristic filtering + dedup?

  mlfoundations/dclm#59 · Filter by author CodeCreator opened on Aug 19, 2024
  9

  

• Dedup methods

  mlfoundations/dclm#54 · Filter by author ch-shin opened on Aug 12, 2024
  1