Datasets:
doc_id string | source string | source_language string | target_language string | target_language_code string | text string | translated_text string | token_count int64 | chunk_count int64 | model string | vllm_server string | created_at string | finalized_at string | chunks_json string | metadata_json string |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
-1000170515976018271 | cosmopedia_v2 | English | Bengali | bn | Superconductivity: A Journey into the World of Zero Resistance and Magnetic Miracles
1. Introduction
In the realm of materials science, there exists a fascinating category known as electronic materials. These substances exhibit properties that make them ideal candidates for various technological applications, particu... | সুপারকন্ডাক্টিভিটি: শূন্য রোধ এবং চৌম্বকীয় বিস্ময়ের জগতে এক যাত্রা
১. ভূমিকা
বস্তুবিজ্ঞানের (materials science) জগতে 'ইলেকট্রনিক ম্যাটেরিয়ালস' নামে একটি চমকপ্রদ বিভাগ রয়েছে। এই পদার্থগুলোর এমন কিছু বৈশিষ্ট্য থাকে যা এদের বিভিন্ন প্রযুক্তিগত প্রয়োগের জন্য, বিশেষ করে ইলেকট্রনিক্স এবং শক্তি রূপান্তরের ক্ষেত্রে আদর্... | 1,092 | 1 | /scratch/shared/oss-models/gemma-4-31B-it | slinky-70:9000 | 2026-06-06T20:44:31.162938+00:00 | 2026-06-08T04:09:34Z | [{"chunk_idx": 0, "char_start": 0, "char_end": 6599, "source_token_count": 1092, "source_text": "Superconductivity: A Journey into the World of Zero Resistance and Magnetic Miracles\n\n1. Introduction\n\nIn the realm of materials science, there exists a fascinating category known as electronic materials. These substanc... | {"all_licenses": null, "arxiv_id": null, "audience": "college_students", "authors": null, "blob_id": null, "branch_name": null, "category": null, "committer_date": null, "config": null, "content_id": null, "dataset_token_count": null, "dataset_token_length": 1344, "detected_licenses": null, "directory_id": null, "exten... |
-1009451830039031165 | cosmopedia_v2 | English | Bengali | bn | ## Understanding Python Data Types and File Manipulation
Python is a versatile programming language that supports various data types and operations on them. In this unit, we will explore different data types, how to manipulate them, and how to work with files using Python.
### Data Types
Python has several built-in ... | ## পাইথনের ডেটা টাইপ এবং ফাইল ম্যানিপুলেশন বোঝা
পাইথন একটি বহুমুখী প্রোগ্রামিং ল্যাঙ্গুয়েজ যা বিভিন্ন ধরনের ডেটা টাইপ এবং সেগুলোর ওপর বিভিন্ন অপারেশন সমর্থন করে। এই ইউনিটে আমরা বিভিন্ন ডেটা টাইপ, সেগুলো কীভাবে ম্যানিপুলেট করতে হয় এবং পাইথন ব্যবহার করে কীভাবে ফাইলের সাথে কাজ করতে হয় তা অন্বেষণ করব।
### ডেটা টাইপ (D... | 1,329 | 1 | /scratch/shared/oss-models/gemma-4-31B-it | slinky-70:9000 | 2026-06-06T22:24:00.515545+00:00 | 2026-06-08T04:09:34Z | [{"chunk_idx": 0, "char_start": 0, "char_end": 4895, "source_token_count": 1329, "source_text": "## Understanding Python Data Types and File Manipulation\n\nPython is a versatile programming language that supports various data types and operations on them. In this unit, we will explore different data types, how to mani... | {"all_licenses": null, "arxiv_id": null, "audience": "college_students", "authors": null, "blob_id": null, "branch_name": null, "category": null, "committer_date": null, "config": null, "content_id": null, "dataset_token_count": null, "dataset_token_length": 1467, "detected_licenses": null, "directory_id": null, "exten... |
-1015927902954080149 | cosmopedia_v2 | English | Bengali | bn | ## Reinforcement Learning: Q-Learning and Softmax Exploration
Reinforcement Learning (RL) is a type of Machine Learning where an agent learns to make decisions by interacting with its environment. The agent receives feedback in the form of rewards or penalties, which it uses to update its knowledge and improve its dec... | ## Reinforcement Learning: Q-Learning এবং Softmax Exploration
Reinforcement Learning (RL) হলো Machine Learning-এর এমন একটি ধরন যেখানে একটি এজেন্ট তার পরিবেশের সাথে মিথস্ক্রিয়ার মাধ্যমে সিদ্ধান্ত নিতে শেখে। এজেন্ট পুরস্কার (rewards) বা দণ্ডের (penalties) আকারে ফিডব্যাক পায়, যা সে তার জ্ঞান আপডেট করতে এবং সিদ্ধান্ত নেও... | 1,088 | 1 | /scratch/shared/oss-models/gemma-4-31B-it | slinky-70:9000 | 2026-06-06T19:28:48.578445+00:00 | 2026-06-08T04:09:34Z | [{"chunk_idx": 0, "char_start": 0, "char_end": 4470, "source_token_count": 1088, "source_text": "## Reinforcement Learning: Q-Learning and Softmax Exploration\n\nReinforcement Learning (RL) is a type of Machine Learning where an agent learns to make decisions by interacting with its environment. The agent receives feed... | {"all_licenses": null, "arxiv_id": null, "audience": "college_students", "authors": null, "blob_id": null, "branch_name": null, "category": null, "committer_date": null, "config": null, "content_id": null, "dataset_token_count": null, "dataset_token_length": 1182, "detected_licenses": null, "directory_id": null, "exten... |
-1019383203153721529 | cosmopedia_v2 | English | Bengali | bn | In many indigenous cultures around the world, midwifery and birthing practices hold great significance and are deeply intertwined with cultural identity, spirituality, and community values. These traditional practices offer valuable insights into holistic approaches to healthcare that center on the wellbeing of both mo... | বিশ্বের অনেক আদিবাসী সংস্কৃতিতে দাইগির কাজ এবং প্রসবকালীন প্রথাগুলোর অত্যন্ত গুরুত্ব রয়েছে এবং এগুলো সাংস্কৃতিক পরিচয়, আধ্যাত্মিকতা এবং সামাজিক মূল্যবোধের সাথে গভীরভাবে মিশে আছে। এই ঐতিহ্যগত প্রথাগুলো স্বাস্থ্যসেবার এমন এক সামগ্রিক দৃষ্টিভঙ্গির মূল্যবান অন্তর্দৃষ্টি প্রদান করে, যা মা এবং শিশু উভয়ের সুস্থতাকে কেন্দ্র... | 1,016 | 1 | /scratch/shared/oss-models/gemma-4-31B-it | slinky-70:9000 | 2026-06-06T19:14:41.570199+00:00 | 2026-06-08T04:09:34Z | [{"chunk_idx": 0, "char_start": 0, "char_end": 6460, "source_token_count": 1016, "source_text": "In many indigenous cultures around the world, midwifery and birthing practices hold great significance and are deeply intertwined with cultural identity, spirituality, and community values. These traditional practices offer... | {"all_licenses": null, "arxiv_id": null, "audience": "college_students", "authors": null, "blob_id": null, "branch_name": null, "category": null, "committer_date": null, "config": null, "content_id": null, "dataset_token_count": null, "dataset_token_length": 1255, "detected_licenses": null, "directory_id": null, "exten... |
-1026969014002256218 | cosmopedia_v2 | English | Bengali | bn | Welcome to our textbook unit on Python programming, where we will explore the fascinating world of coding! Today, we will dive into a Python code snippet from a tutorial and learn about some essential concepts along the way. Let's begin our journey!
Firstly, let's ensure we have the necessary libraries installed. For ... | পাইথন প্রোগ্রামিং-এর পাঠ্যবই ইউনিটে আপনাদের স্বাগতম, যেখানে আমরা কোডিং-এর চমৎকার জগৎটি অন্বেষণ করব! আজ আমরা একটি টিউটোরিয়াল থেকে পাইথন কোড স্নিপেট নিয়ে আলোচনা করব এবং সেই সাথে কিছু প্রয়োজনীয় ধারণা সম্পর্কে জানব। চলুন আমাদের যাত্রা শুরু করা যাক!
প্রথমত, আমাদের প্রয়োজনীয় লাইব্রেরিগুলো ইনস্টল করা আছে কি না তা নিশ্চিত কর... | 1,166 | 1 | /scratch/shared/oss-models/gemma-4-31B-it | slinky-70:9000 | 2026-06-06T23:21:11.336110+00:00 | 2026-06-08T04:09:34Z | [{"chunk_idx": 0, "char_start": 0, "char_end": 4955, "source_token_count": 1166, "source_text": "Welcome to our textbook unit on Python programming, where we will explore the fascinating world of coding! Today, we will dive into a Python code snippet from a tutorial and learn about some essential concepts along the way... | {"all_licenses": null, "arxiv_id": null, "audience": "college_students", "authors": null, "blob_id": null, "branch_name": null, "category": null, "committer_date": null, "config": null, "content_id": null, "dataset_token_count": null, "dataset_token_length": 1295, "detected_licenses": null, "directory_id": null, "exten... |
-1041099932798805265 | cosmopedia_v2 | English | Bengali | bn | "## Linear Regression: Predicting Car Stopping Distances\n\nIn this unit, we will explore the concep(...TRUNCATED) | "## লিনিয়ার রিগ্রেশন: গাড়ির থামার দূর(...TRUNCATED) | 1,191 | 1 | /scratch/shared/oss-models/gemma-4-31B-it | slinky-70:9000 | 2026-06-06T20:05:59.198669+00:00 | 2026-06-08T04:09:34Z | "[{\"chunk_idx\": 0, \"char_start\": 0, \"char_end\": 4621, \"source_token_count\": 1191, \"source_t(...TRUNCATED) | "{\"all_licenses\": null, \"arxiv_id\": null, \"audience\": \"college_students\", \"authors\": null,(...TRUNCATED) |
-1043361505298524071 | cosmopedia_v2 | English | Bengali | bn | "Course Unit: Muslim Women in Soviet Visual Propaganda during the Interwar Period\n\nIn this course (...TRUNCATED) | "কোর্স ইউনিট: আন্তর যুদ্ধকালীন সময়ে সো(...TRUNCATED) | 1,149 | 1 | /scratch/shared/oss-models/gemma-4-31B-it | slinky-70:9000 | 2026-06-06T23:03:11.383829+00:00 | 2026-06-08T04:09:34Z | "[{\"chunk_idx\": 0, \"char_start\": 0, \"char_end\": 7259, \"source_token_count\": 1149, \"source_t(...TRUNCATED) | "{\"all_licenses\": null, \"arxiv_id\": null, \"audience\": \"college_students\", \"authors\": null,(...TRUNCATED) |
-1048435489483062020 | cosmopedia_v2 | English | Bengali | bn | "Chapter 10: Wittgensteinian Language Games and Moral Philosophy's Crisis\n\nIntroduction\n\nLudwig (...TRUNCATED) | "অধ্যায় ১০: উইটগেনস্টাইনীয় ল্যাঙ্গু(...TRUNCATED) | 1,249 | 1 | /scratch/shared/oss-models/gemma-4-31B-it | slinky-70:9000 | 2026-06-06T21:38:13.499813+00:00 | 2026-06-08T04:09:34Z | "[{\"chunk_idx\": 0, \"char_start\": 0, \"char_end\": 7553, \"source_token_count\": 1249, \"source_t(...TRUNCATED) | "{\"all_licenses\": null, \"arxiv_id\": null, \"audience\": \"college_students\", \"authors\": null,(...TRUNCATED) |
-1051977906166232076 | cosmopedia_v2 | English | Bengali | bn | "In the vast and ever-evolving landscape of global entertainment, few phenomena have captured the wo(...TRUNCATED) | "বিশ্ব বিনোদনের বিশাল এবং প্রতিনিয়ত প(...TRUNCATED) | 1,167 | 1 | /scratch/shared/oss-models/gemma-4-31B-it | slinky-70:9000 | 2026-06-06T21:31:43.281048+00:00 | 2026-06-08T04:09:34Z | "[{\"chunk_idx\": 0, \"char_start\": 0, \"char_end\": 6114, \"source_token_count\": 1167, \"source_t(...TRUNCATED) | "{\"all_licenses\": null, \"arxiv_id\": null, \"audience\": \"college_students\", \"authors\": null,(...TRUNCATED) |
-1056851178325763512 | cosmopedia_v2 | English | Bengali | bn | "In the realm of game development, the theoretical framework of constructivism offers us a unique le(...TRUNCATED) | "গেম ডেভেলপমেন্টের ক্ষেত্রে, কনস্ট্র(...TRUNCATED) | 1,287 | 1 | /scratch/shared/oss-models/gemma-4-31B-it | slinky-70:9000 | 2026-06-06T22:14:12.687264+00:00 | 2026-06-08T04:09:34Z | "[{\"chunk_idx\": 0, \"char_start\": 0, \"char_end\": 7778, \"source_token_count\": 1287, \"source_t(...TRUNCATED) | "{\"all_licenses\": null, \"arxiv_id\": null, \"audience\": \"college_students\", \"authors\": null,(...TRUNCATED) |
End of preview. Expand in Data Studio
C3PO Mixed Pretraining Corpus Translations
English → Indian language translations of sumanthd/c3po-mixed-pretraining-corpus.
Each language is a separate dataset config (subset) on the Hub for easier browsing.
Documents are translated with Gemma 4 31B via vLLM in ~4k-token chunks at
paragraph/sentence boundaries, then stitched into full documents. Individual
chunk translations are preserved in chunks_json.
Stats
- Total rows: 145,932
- Source corpus:
sumanthd/c3po-mixed-pretraining-corpus - Model:
/scratch/shared/oss-models/gemma-4-31B-it - Published: 2026-06-08T16:27:56+00:00
Configs (one per language)
| Config | Language | Documents |
|---|---|---|
bn |
Bengali | 24,322 |
hi |
Hindi | 24,322 |
kn |
Kannada | 24,322 |
ml |
Malayalam | 24,322 |
ta |
Tamil | 24,322 |
te |
Telugu | 24,322 |
Partial languages will be added as new configs when translation completes.
Schema
| Field | Type | Description |
|---|---|---|
doc_id |
string | Source document id |
source |
string | Source slug (wikipedia, proof_pile, …) |
source_language |
string | Always English |
target_language |
string | Target language name |
target_language_code |
string | ISO 639-1 code |
text |
string | English source document |
translated_text |
string | Stitched full translation |
chunks_json |
string | JSON array of chunk metadata (source + translated text, char spans) |
metadata_json |
string | JSON object of source-specific metadata |
token_count |
int | Source token count (Gemma tokenizer) |
chunk_count |
int | Number of translation chunks |
model |
string | Translation model id |
vllm_server |
string | vLLM router used |
created_at |
string | Translation timestamp (UTC) |
finalized_at |
string | Final stitch timestamp (UTC) |
Load
from datasets import load_dataset
import json
# Pick a language config
bn = load_dataset("sumanthd/c3po-mixed-pretraining-corpus-translations", "bn", split="train")
hi = load_dataset("sumanthd/c3po-mixed-pretraining-corpus-translations", "hi", split="train")
kn = load_dataset("sumanthd/c3po-mixed-pretraining-corpus-translations", "kn", split="train")
ml = load_dataset("sumanthd/c3po-mixed-pretraining-corpus-translations", "ml", split="train")
ta = load_dataset("sumanthd/c3po-mixed-pretraining-corpus-translations", "ta", split="train")
te = load_dataset("sumanthd/c3po-mixed-pretraining-corpus-translations", "te", split="train")
print(hi[0]["translated_text"][:200])
chunks = json.loads(hi[0]["chunks_json"])
print(chunks[0].keys())
- Downloads last month
- 6