Add agg_Latn shard 0: 33002 sentences, 934 paragraphs, 27317 words, 759655 ngrams

README.md

@@ -947,6 +947,22 @@ configs:
     data_files:
       - split: train
         path: data/ngrams/kmg_Latn/*.parquet
+  - config_name: sentences-agg_Latn
+    data_files:
+      - split: train
+        path: data/sentences/agg_Latn/*.parquet
+  - config_name: paragraphs-agg_Latn
+    data_files:
+      - split: train
+        path: data/paragraphs/agg_Latn/*.parquet
+  - config_name: words-agg_Latn
+    data_files:
+      - split: train
+        path: data/words/agg_Latn/*.parquet
+  - config_name: ngrams-agg_Latn
+    data_files:
+      - split: train
+        path: data/ngrams/agg_Latn/*.parquet
   - config_name: sentences-guj_Latn
     data_files:
       - split: train
@@ -979,27 +995,11 @@ configs:
     data_files:
       - split: train
         path: data/ngrams/knv_Latn/*.parquet
-  - config_name: sentences-dgz_Latn
-    data_files:
-      - split: train
-        path: data/sentences/dgz_Latn/*.parquet
-  - config_name: paragraphs-dgz_Latn
-    data_files:
-      - split: train
-        path: data/paragraphs/dgz_Latn/*.parquet
-  - config_name: words-dgz_Latn
-    data_files:
-      - split: train
-        path: data/words/dgz_Latn/*.parquet
-  - config_name: ngrams-dgz_Latn
-    data_files:
-      - split: train
-        path: data/ngrams/dgz_Latn/*.parquet
 ---
 
 # FineWeb-2 NLP
 
-**12,608,974 sentences** across **819 languages**, extracted from **256,960 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.
+**12,641,976 sentences** across **820 languages**, extracted from **257,894 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.
 
 ## Table of Contents
 
@@ -1036,10 +1036,10 @@ linguistically useful units:
 
 | Type | Rows | What you get |
 |------|------|-------------|
-| **sentences** | 12,608,974 | One row per sentence, with source document ID, URL, and position index |
-| **paragraphs** | 262,555 | One row per paragraph, with sentence count per paragraph |
-| **words** | 11,164,936 | Per-shard word frequency and document frequency tables |
-| **ngrams** | 371,693,137 | Per-shard bigram through 5-gram frequency tables |
+| **sentences** | 12,641,976 | One row per sentence, with source document ID, URL, and position index |
+| **paragraphs** | 263,489 | One row per paragraph, with sentence count per paragraph |
+| **words** | 11,192,253 | Per-shard word frequency and document frequency tables |
+| **ngrams** | 372,452,792 | Per-shard bigram through 5-gram frequency tables |
 
 Every row traces back to its source document through `doc_id` and `doc_url` fields, making
 it possible to navigate from any sentence or word back to the original web page. This
@@ -1194,6 +1194,7 @@ cuk_Latn       ▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓
 mgh_Latn       ▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓ 997
 pnt_Grek       ▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓ 946
 cui_Latn       ▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓ 934
+agg_Latn       ▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓ 934
 dop_Latn       ▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓ 925
 ncx_Latn       ▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓ 923
 shi_Latn       ▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓ 917
@@ -1207,7 +1208,6 @@ qwh_Latn       ▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓
 dar_Cyrl       ▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓ 886
 koo_Latn       ▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓ 881
 kmr_Cyrl       ▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓ 876
-dik_Latn       ▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓ 873
 ```
 
 <details>
@@ -1368,14 +1368,14 @@ print(words_df)
 
 | Metric | Value |
 |--------|-------|
-| **Total sentences** | **12,608,974** |
-| **Total paragraphs** | **262,555** |
-| **Total word entries** (per-shard) | **11,164,936** |
-| **Total n-gram entries** (per-shard) | **371,693,137** |
-| **Languages processed** | **819** |
-| **Source documents** | **256,960** |
+| **Total sentences** | **12,641,976** |
+| **Total paragraphs** | **263,489** |
+| **Total word entries** (per-shard) | **11,192,253** |
+| **Total n-gram entries** (per-shard) | **372,452,792** |
+| **Languages processed** | **820** |
+| **Source documents** | **257,894** |
 | **Total parquet size** | **4.3 GB** |
-| Avg sentences per document | 49.1 |
+| Avg sentences per document | 49.0 |
 | Avg paragraphs per document | 1.0 |
 | Avg sentences per paragraph | 48.0 |
 
@@ -1440,9 +1440,9 @@ print(words_df)
 | 55 | `cbu_Latn` | 33,179 | 402 | 32,390 | 802,729 | 402 | 1 | 11.6 MB |
 | 56 | `des_Latn` | 33,174 | 857 | 17,121 | 694,481 | 857 | 1 | 8.5 MB |
 | 57 | `kmg_Latn` | 33,056 | 677 | 29,983 | 317,466 | 677 | 1 | 5.3 MB |
-| 58 | `guj_Latn` | 32,948 | 756 | 80,680 | 643,977 | 713 | 1 | 9.8 MB |
-| 59 | `knv_Latn` | 32,816 | 549 | 23,525 | 663,975 | 549 | 1 | 9.1 MB |
-| 60 | `dgz_Latn` | 32,778 | 433 | 11,509 | 720,150 | 433 | 1 | 7.8 MB |
+| 58 | `agg_Latn` | 33,002 | 934 | 27,317 | 759,655 | 934 | 1 | 11.6 MB |
+| 59 | `guj_Latn` | 32,948 | 756 | 80,680 | 643,977 | 713 | 1 | 9.8 MB |
+| 60 | `knv_Latn` | 32,816 | 549 | 23,525 | 663,975 | 549 | 1 | 9.1 MB |
 
 ## How it works
 
@@ -1713,4 +1713,4 @@ Created by **Duc-Tam Nguyen** ([tamnd](https://huggingface.co/tamnd)) as part of
 
 ---
 
-*Last updated: 2026-04-15 01:39 UTC*
+*Last updated: 2026-04-15 01:40 UTC*

data/ngrams/agg_Latn/0000.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5fef61993f13e75b12761b52f707893760d493b7ee7b32c97c8a212ec31efce0
+size 9446941

data/paragraphs/agg_Latn/0000.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:931cd475c38d3eaa2128aff175724f212f5ddd2bf9eb2836623b98fa66d7344f
+size 1407363

data/sentences/agg_Latn/0000.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ecf855c14bed82c4ed7696f1372ad095666dd2f12f0c67b3dab8e2be903439ea
+size 1107386

data/words/agg_Latn/0000.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c2fb5cf9582ef35ebd1e9d0042b9e9c2f2e517de618478c6c52a2f52e35c7b2a
+size 170624

stats.csv

@@ -818,3 +818,4 @@ kpg_Latn,kpg,0,913,43245,914,9360,849682,10131488,2026-04-15T01:38:48Z
 ncx_Latn,ncx,0,919,46103,923,34531,468172,8415933,2026-04-15T01:39:12Z
 dop_Latn,dop,0,924,35229,925,18770,295365,4871709,2026-04-15T01:39:34Z
 cui_Latn,cui,0,934,18105,934,12811,484019,6168579,2026-04-15T01:39:56Z
+agg_Latn,agg,0,934,33002,934,27317,759655,12132314,2026-04-15T01:40:21Z