{
  "counts_by_dataset_source": {
    "git10m": 6392684,
    "rsteller": 1309707
  },
  "counts_by_gsd_group": {
    "2m": 827031,
    "4m": 569806,
    "8m": 267935,
    "le1": 6037619
  },
  "filter_policy": {
    "dataset_source_values": [
      "git10m",
      "rsteller"
    ],
    "git10m_gsd_max_m_per_pixel": 8.0,
    "git_token_count_source": "caption_length",
    "min_caption_tokens": 20,
    "rsteller_token_count_source": "/data/users/yangcong/Qwen/Qwen3.5-0.8B-Base"
  },
  "inputs": {
    "git_dir": "/data/users/yangcong/UniRSDATA/Git-10M",
    "rsteller_dir": "/data/users/yangcong/UniRSDATA/RSTeller",
    "tokenizer_path": "/data/users/yangcong/Qwen/Qwen3.5-0.8B-Base"
  },
  "output": {
    "format": "One Hugging Face DatasetDict with one sharded Apache Arrow train split",
    "max_shard_size": "2GB",
    "path": "/data/users/yangcong/UniRSDATA/Filtered-Pool-Git10M-le8-RSTeller-min20-Arrow",
    "rows": 7702391,
    "sources_are_merged_not_separate": true
  },
  "ratios_by_dataset_source": {
    "git10m": 0.829960982245643,
    "rsteller": 0.17003901775435706
  },
  "ratios_by_gsd_group": {
    "2m": 0.10737328188091204,
    "4m": 0.07397780767037145,
    "8m": 0.034785951531154415,
    "le1": 0.7838629589175621
  }
}