avey-b-snd / MANIFEST.json
DanishMahdi's picture
Publish Avey-B Sindhi 25M
a9088e0 verified
Raw History Blame Contribute Delete
1.35 kB
{
"generated_at": "2026-09-11T07:23:51.120123+00:00",
"lang": "snd_Arab",
"map_url": "https://data.hplt-project.org/three/sorted/snd_Arab.map",
"bins_requested": [
10,
9
],
"prob_threshold": 0.8,
"min_chars": 20,
"val_fraction": 0.005,
"max_docs_per_shard": 50000,
"max_chars_per_shard": 500000000,
"max_shards": null,
"stream_download": true,
"shards": [
{
"url": "https://data.hplt-project.org/three/sorted/snd_Arab/10_1.jsonl.zst",
"bin": 10,
"raw_docs": 167,
"dropped_low_prob": 0,
"dropped_empty": 0,
"dropped_exact_dup": 0,
"pii_spans_redacted": 17,
"kept_train": 167,
"kept_val": 0,
"chars_kept": 3752256
},
{
"url": "https://data.hplt-project.org/three/sorted/snd_Arab/9_1.jsonl.zst",
"bin": 9,
"raw_docs": 32166,
"dropped_low_prob": 9,
"dropped_empty": 0,
"dropped_exact_dup": 0,
"pii_spans_redacted": 1356,
"kept_train": 32004,
"kept_val": 153,
"chars_kept": 247251424
}
],
"totals": {
"raw_docs": 32333,
"kept_train": 32171,
"kept_val": 153,
"dropped_low_prob": 9,
"dropped_empty": 0,
"dropped_exact_dup": 0,
"pii_spans_redacted": 1373,
"chars_kept": 251003680
},
"consolidated_counts": {
"train": 32171,
"val": 153
}
}