tamnd commited on
Commit
a9fdf8c
·
verified ·
1 Parent(s): ef532eb

Add agr_Latn shard 0: 19136 sentences, 463 paragraphs, 34832 words, 725589 ngrams

Browse files
README.md CHANGED
@@ -999,7 +999,7 @@ configs:
999
 
1000
  # FineWeb-2 NLP
1001
 
1002
- **7,400,841 sentences** across **628 languages**, extracted from **135,642 source documents** in [FineWeb-2](https://huggingface.co/datasets/HuggingFaceFW/fineweb-2). Every sentence, paragraph, word frequency, and n-gram frequency, split with language-aware segmentation and continuously updated.
1003
 
1004
  ## Table of Contents
1005
 
@@ -1036,10 +1036,10 @@ linguistically useful units:
1036
 
1037
  | Type | Rows | What you get |
1038
  |------|------|-------------|
1039
- | **sentences** | 7,400,841 | One row per sentence, with source document ID, URL, and position index |
1040
- | **paragraphs** | 138,790 | One row per paragraph, with sentence count per paragraph |
1041
- | **words** | 6,889,166 | Per-shard word frequency and document frequency tables |
1042
- | **ngrams** | 260,031,921 | Per-shard bigram through 5-gram frequency tables |
1043
 
1044
  Every row traces back to its source document through `doc_id` and `doc_url` fields, making
1045
  it possible to navigate from any sentence or word back to the original web page. This
@@ -1193,6 +1193,7 @@ mgh_Latn ▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓
1193
  ikt_Latn ▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓ 516
1194
  rmy_Cyrl ▓▓▓▓▓▓▓▓▓▓▓▓▓▓ 485
1195
  smk_Latn ▓▓▓▓▓▓▓▓▓▓▓▓▓▓ 476
 
1196
  guc_Latn ▓▓▓▓▓▓▓▓▓▓▓▓▓ 462
1197
  enm_Latn ▓▓▓▓▓▓▓▓▓▓▓▓▓ 461
1198
  gof_Latn ▓▓▓▓▓▓▓▓▓▓▓▓▓ 460
@@ -1202,12 +1203,11 @@ snd_Latn ▓▓▓▓▓▓▓▓▓▓▓▓▓ 454
1202
  ozm_Latn ▓▓▓▓▓▓▓▓▓▓▓▓▓ 453
1203
  gde_Latn ▓▓▓▓▓▓▓▓▓▓▓▓▓ 453
1204
  dad_Latn ▓▓▓▓▓▓▓▓▓▓▓▓▓ 451
1205
- pwg_Latn ▓▓▓▓▓▓▓▓▓▓▓▓▓ 448
1206
  tee_Latn ▓▓▓▓▓▓▓▓▓▓▓▓▓ 448
 
1207
  bmu_Latn ▓▓▓▓▓▓▓▓▓▓▓▓▓ 447
1208
  cax_Latn ▓▓▓▓▓▓▓▓▓▓▓▓▓ 447
1209
- taq_Latn ▓▓▓▓▓▓▓▓▓▓▓▓▓ 446
1210
- tpt_Latn ▓▓▓▓▓▓▓▓▓▓▓▓▓ 446
1211
  ```
1212
 
1213
  <details>
@@ -1368,14 +1368,14 @@ print(words_df)
1368
 
1369
  | Metric | Value |
1370
  |--------|-------|
1371
- | **Total sentences** | **7,400,841** |
1372
- | **Total paragraphs** | **138,790** |
1373
- | **Total word entries** (per-shard) | **6,889,166** |
1374
- | **Total n-gram entries** (per-shard) | **260,031,921** |
1375
- | **Languages processed** | **628** |
1376
- | **Source documents** | **135,642** |
1377
  | **Total parquet size** | **2.9 GB** |
1378
- | Avg sentences per document | 54.6 |
1379
  | Avg paragraphs per document | 1.0 |
1380
  | Avg sentences per paragraph | 53.3 |
1381
 
 
999
 
1000
  # FineWeb-2 NLP
1001
 
1002
+ **7,419,977 sentences** across **629 languages**, extracted from **136,104 source documents** in [FineWeb-2](https://huggingface.co/datasets/HuggingFaceFW/fineweb-2). Every sentence, paragraph, word frequency, and n-gram frequency, split with language-aware segmentation and continuously updated.
1003
 
1004
  ## Table of Contents
1005
 
 
1036
 
1037
  | Type | Rows | What you get |
1038
  |------|------|-------------|
1039
+ | **sentences** | 7,419,977 | One row per sentence, with source document ID, URL, and position index |
1040
+ | **paragraphs** | 139,253 | One row per paragraph, with sentence count per paragraph |
1041
+ | **words** | 6,923,998 | Per-shard word frequency and document frequency tables |
1042
+ | **ngrams** | 260,757,510 | Per-shard bigram through 5-gram frequency tables |
1043
 
1044
  Every row traces back to its source document through `doc_id` and `doc_url` fields, making
1045
  it possible to navigate from any sentence or word back to the original web page. This
 
1193
  ikt_Latn ▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓ 516
1194
  rmy_Cyrl ▓▓▓▓▓▓▓▓▓▓▓▓▓▓ 485
1195
  smk_Latn ▓▓▓▓▓▓▓▓▓▓▓▓▓▓ 476
1196
+ agr_Latn ▓▓▓▓▓▓▓▓▓▓▓▓▓ 463
1197
  guc_Latn ▓▓▓▓▓▓▓▓▓▓▓▓▓ 462
1198
  enm_Latn ▓▓▓▓▓▓▓▓▓▓▓▓▓ 461
1199
  gof_Latn ▓▓▓▓▓▓▓▓▓▓▓▓▓ 460
 
1203
  ozm_Latn ▓▓▓▓▓▓▓▓▓▓▓▓▓ 453
1204
  gde_Latn ▓▓▓▓▓▓▓▓▓▓▓▓▓ 453
1205
  dad_Latn ▓▓▓▓▓▓▓▓▓▓▓▓▓ 451
 
1206
  tee_Latn ▓▓▓▓▓▓▓▓▓▓▓▓▓ 448
1207
+ pwg_Latn ▓▓▓▓▓▓▓▓▓▓▓▓▓ 448
1208
  bmu_Latn ▓▓▓▓▓▓▓▓▓▓▓▓▓ 447
1209
  cax_Latn ▓▓▓▓▓▓▓▓▓▓▓▓▓ 447
1210
+ bhl_Latn ▓▓▓▓▓▓▓▓▓▓▓▓▓ 446
 
1211
  ```
1212
 
1213
  <details>
 
1368
 
1369
  | Metric | Value |
1370
  |--------|-------|
1371
+ | **Total sentences** | **7,419,977** |
1372
+ | **Total paragraphs** | **139,253** |
1373
+ | **Total word entries** (per-shard) | **6,923,998** |
1374
+ | **Total n-gram entries** (per-shard) | **260,757,510** |
1375
+ | **Languages processed** | **629** |
1376
+ | **Source documents** | **136,104** |
1377
  | **Total parquet size** | **2.9 GB** |
1378
+ | Avg sentences per document | 54.5 |
1379
  | Avg paragraphs per document | 1.0 |
1380
  | Avg sentences per paragraph | 53.3 |
1381
 
data/ngrams/agr_Latn/0000.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0a692035da08fa59a7ec77bd68ebd0580f37409e577d02117be1a52ea33a0c4b
3
+ size 8829821
data/paragraphs/agr_Latn/0000.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5b1ed8cf065c6370a4df1944f110509a953cf8d4b01cfa4229d73ec2e6f6d215
3
+ size 862335
data/sentences/agr_Latn/0000.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:634e378f801b0caad423b7244adf4de75ada44981648951b6aeec23f2c7527aa
3
+ size 782440
data/words/agr_Latn/0000.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1bc1c2b228171c5a05c0e42bd693f253d628e746da026aee7c8012953d30246f
3
+ size 180188
stats.csv CHANGED
@@ -627,3 +627,4 @@ thk_Latn,thk,0,457,20249,457,23679,555060,7882442,2026-04-15T00:01:23Z
627
  gde_Latn,gde,0,453,18353,453,9462,615717,7430879,2026-04-15T00:01:52Z
628
  pah_Latn,pah,0,456,22573,456,14188,539478,6446792,2026-04-15T00:02:25Z
629
  gof_Latn,gof,0,460,21073,460,21833,564206,7946417,2026-04-15T00:03:01Z
 
 
627
  gde_Latn,gde,0,453,18353,453,9462,615717,7430879,2026-04-15T00:01:52Z
628
  pah_Latn,pah,0,456,22573,456,14188,539478,6446792,2026-04-15T00:02:25Z
629
  gof_Latn,gof,0,460,21073,460,21833,564206,7946417,2026-04-15T00:03:01Z
630
+ agr_Latn,agr,0,462,19136,463,34832,725589,10654784,2026-04-15T00:03:33Z