Update README.md

README.md

@@ -1,8 +1,36 @@
 ---
 license: odc-by
+pretty_name: Zyda2-5T
+task_categories:
+- text-generation
+language:
+- en
+size_categories:
+- n>1T
+configs:
+  - config_name: default
+    data_files:
+      - split: train
+        path: data/*/*/*
+  - config_name: dclm-crossdeduped
+    data_files:
+      - split: train
+        path: data/dclm-baseline/*/*
+  - config_name: zyda-crossdeduped-filtered 
+    data_files:
+      - split: train
+        path: data/zyda/*/*
+  - config_name: dolma_cc-crossdeduped-filtered
+    data_files:
+      - split: train
+        path: data/dolma-cc-v1_7-deduped-filtered/*
+  - config_name: fwe3
+    data_files:
+      - split: train
+        path: data/fwe3/*/*
 ---
 
-# Zynemo-5T
+# Zyda2-5T
 
 <!-- Provide a quick summary of the dataset. -->
 
@@ -24,12 +52,31 @@ According to our evaluations, Zynemo is the most performant per-token open datas
 For more information, please see our technical blog (-/TODO LINK)
 
 ## How to download
+Since we preserved the schemas of original component datasets, attempting to dowlnoad the whole dataset using `datasets.load_dataset()` might fail during the stage of generating a split.
+
+To download the whole dataset we recommend to either clone the repository, or, if you must use the `datasets.load_dataset()`, then download individual components separately.
+
+Example command to clone the repository using huggingface-cli: `huggingface-cli download Zyphra/Zyda2-5T--repo-type dataset`
+
+Commands to download individual components:
+ - DCLM: `ds = datasets.load_dataset("Zyphra/Zyda2-5T", name="dclm-crossdeduped", split="train")`
+ - Zyda: `ds = datasets.load_dataset("Zyphra/Zyda2-5T", name="zyda-crossdeduped-filtered ", split="train")`
+ - Dolma-CC: `ds = datasets.load_dataset("Zyphra/Zyda2-5T", name="dolma_cc-crossdeduped-filtered", split="train")`
+ - Fineweb-Edu: `ds = datasets.load_dataset("Zyphra/Zyda2-5T", name="fwe3", split="train")`
+
+In this repository we provide raw results of cross deduplication and filtering. To achieve the best possible performance, one will need to appropriate weights during training.
+We found the following optimal weights (in the sense of weights in the resultant dataset): DCLM - 4.0, FWE3 - 4.0, Zyda - 0.16, Dolma-CC - 0.24.
 
-// TODO YURY
 
 ## Breakdown by component
 
-// TODO YURY
+| Component                    | Download size (parquet, GBs) | Documents (millions) | gpt-neox tokens (billions) |
+| --- | --- | --- | --- |
+| dclm-crossdeduped              |   8469.4                   |   2,590.5            |   3,348.942                |
+| zyda-crossdeduped-filtered     |   452.4                    |   247.7              |   163.6                    |
+| dolma_cc-crossdeduped-filtered |   668.2                    |   445.6              |   238.4                    |
+| fwe3                           |   3490.5                   |   1,279.1            |   1,319.2                  |
+| Total                          |   13080.5                  |   4,562.8            |   5,070.2                  |
 
 ### Dataset Description
 
@@ -44,18 +91,15 @@ For more information, please see our technical blog (-/TODO LINK)
 
 <!-- This section provides a description of the dataset fields, and additional information about the dataset structure such as criteria used to create the splits, relationships between data points, etc. -->
 
-// TODO IS THIS CORRECT YURY?
+Each component has they're own individual schema. Please, consult with their respective sources for exact information.
 
+However, in all components the document text is in `text` column, and unique document document id is in `nemo_id` column.
 
-Dataset fields:
-- `text`: contains actual text for training
-- `source`: component the text is coming from
-- `filtering_features`: precomputed values of different features that were used for filtering (converted to json string)
-- `source_other`: metadata from the source dataset (converted to json string)
+Our Zyda1 and Dolma-CC versions also have two additional columns corresponding to prediction of Nvidia's quality model (https://huggingface.co/nvidia/quality-classifier-deberta): `quality_prob` and `quality_pred`. 
 
 ### Source Data
 
-Zynemo is comprised of four high quality open-source datasets:
+Zyda2 is comprised of four high quality open-source datasets:
 
 Zyda1: https://huggingface.co/datasets/Zyphra/Zyda
 
@@ -63,7 +107,7 @@ Dolma-1.7-cc https://huggingface.co/datasets/allenai/dolma
 
 DCLM-baseline: https://huggingface.co/datasets/mlfoundations/dclm-baseline-1.0
 
-FineWeb-Edu-2 https://huggingface.co/datasets/HuggingFaceFW/fineweb-edu
+FineWeb-Edu-score2: https://huggingface.co/datasets/HuggingFaceFW/fineweb-edu-score-2
 
 <center>
 <img src="https://cdn-uploads.huggingface.co/production/uploads/65c05e75c084467acab2f84a/GQenkNxzyM65M4eR2YZcV.png" width="600" alt="ZyNeMo dataset composition">