datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
corpus-1T-manifest
SPP Corpus 1T Manifest
The selection manifest for the ~1.0T-token pretraining corpus used in
Synthetic Persona Pretraining (SPP): Alignment from Token Zero.
The corpus is a seeded subsample of allenai/dolma3_mix-6T.
Rather than redistribute ~2.6 TB of text that is already public, this dataset
publishes the selection decisions keyed by upstream document id, so the corpus
can be reconstructed exactly by replaying against upstream.
📄 Reflections + text for the annotated half:… See the full description on the dataset page: https://huggingface.co/datasets/dlab-spp/corpus-1T-manifest.PulseLM
PulseLM: A Foundation Dataset and Benchmark for PPG-Text Learning
Usage
from datasets import load_dataset, get_dataset_config_names, concatenate_datasets # datasets==4.5.0
dataset_names = get_dataset_config_names("Manhph2211/PulseLM")
print(f"Available datasets: {dataset_names}")
train_splits = [
load_dataset("Manhph2211/PulseLM", name, split="train").select_columns(["signal", "text", "qa"])
for name in dataset_names
]
combined =… See the full description on the dataset page: https://huggingface.co/datasets/Manhph2211/PulseLM.Omni-Frontier-Distillation-SFT-Cyber-security-Coding-dataset-collection-v2
🧬 Omni-Frontier Collection
Cybersecurity · Coding · Math · Science · RSI Reasoning — one unified SFT package
A unified, deduplicated, fully-browsable distillation & SFT corpus — every row real, every row visible.
📖 Jump to
What's inside · 🔁 Aggregation audit · 🛡 Cybersecurity · 💻 Coding · 🏭 Distillation deep-dive · 🔁 RSI · 🧮 Math/Science/More · 🎓 Training guide · 🔎 Browsing · 🧹 Quality · 🗺 Roadmap · 📄 License… See the full description on the dataset page: https://huggingface.co/datasets/Manusagents/Omni-Frontier-Distillation-SFT-Cyber-security-Coding-dataset-collection-v2.python-copilot-training-from-many-repos-large
Python Copilot Large Coding Dataset
This dataset is a subset of the matlok python copilot datasets. Please refer to the Multimodal Python Copilot Training Overview for more details on how to use this dataset.
Details
Each row contains python code, either a class method or a global function, imported modules, base classes (if any), exceptions (ordered based off the code), returns (ordered based off the code), arguments (ordered based off the code), and more.
Rows: 2350782… See the full description on the dataset page: https://huggingface.co/datasets/matlok/python-copilot-training-from-many-repos-large.ManipuriGPT-Corpus-v1.0
ManipuriGPT Corpus v1.0
ManipuriGPT Corpus v1.0 is a research-grade, multi-script, deduplicated, and quality-scored corpus specifically engineered for pretraining Manipuri (Meiteilon) language foundation models.
Quick Summary
Total Sequences: 147,956
Total Tokens (ManipuriGPT-Tokenizer-v1.0): 4,347,075
Total Characters: 16,019,401
Pipeline Version: 5.6
Release Version: v1.0.0
Build Timestamp: 2026-07-25T09:17:21.960438Z
Primary Writing Systems… See the full description on the dataset page: https://huggingface.co/datasets/nanskong/ManipuriGPT-Corpus-v1.0.GPT-5.6-Sol-Luna-Terra-Traces
GPT-5.6 — Sol · Terra · Luna Library
A maintained mirror of every GPT-5.6 Sol / Terra / Luna dataset on Hugging Face — content-verified, attributed, in one place.
Dataset Viewer | Parquet
// what this is
This is a maintained library — a community mirror of every publicly-available GPT-5.6 Sol / Terra / Luna dataset on Hugging Face, aggregated, validity-filtered, and content-verified with per-row source attribution. It is not Crownelius' own data. It exists to… See the full description on the dataset page: https://huggingface.co/datasets/Manusagents/GPT-5.6-Sol-Luna-Terra-Traces.autoresearch-manim
Autoresearch Manim
Curated Manim code-generation examples exported from the autoresearch_manim_finetune pipeline.
Preview Gallery
Preview
Preview
Preview
Machine learning: attention plus residual mixing
Physics: boundary layer flow near a surface
Biology: neuron structure and signal direction
Finance: compound growth over time
Economics: production frontier tradeoff
Neuroscience: action potential phases
Summary
Focus:… See the full description on the dataset page: https://huggingface.co/datasets/sebastianboehler/autoresearch-manim.historical-training-manuals
Historical Training Manuals
1,597 US government and government-adjacent training manuals and technical publications
sourced from the Internet Archive, spanning roughly 1800-2021. Records carry
bibliographic metadata; a subset also carries extracted full text and a machine-generated
summary.
Loading
from datasets import load_dataset
ds = load_dataset("robworks-software/historical-training-manuals")
Splits
Split
Rows
train
1,277… See the full description on the dataset page: https://huggingface.co/datasets/robworks-software/historical-training-manuals.kotlin-test-pairs
KotlinTestPairs
9,856 Kotlin source↔unit-test pairs ("focal method" pairs) mined from permissively
licensed public GitHub code.
Applies the methods2test (MSR 2022) methodology to
Kotlin, and is the sibling of
SwiftTestPairs.
⚠️ This dataset contains NO source code
Rows are references + derived metadata: repository, file paths, content MD5s, and
measured properties. The code stays where it has always been, in the upstream corpus.
This avoids redistributing anyone's… See the full description on the dataset page: https://huggingface.co/datasets/Manju46/kotlin-test-pairs.swift-test-pairs
SwiftTestPairs
20,801 Swift source↔unit-test pairs ("focal method" pairs) mined from permissively
licensed public GitHub code.
Applies the methods2test (MSR 2022) methodology to
Swift, which had no equivalent dataset.
⚠️ This dataset contains NO source code
Rows are references + derived metadata: repository, file paths, content MD5s, and
measured properties. Reconstruct file contents from the two public upstream datasets with resolve.py
(⚠️ streams ~4 GB from the… See the full description on the dataset page: https://huggingface.co/datasets/Manju46/swift-test-pairs.ManikKatha
ManikKatha — Manik Bandopadhyay Bengali Literary Corpus
Bengali literary prose by Manik Bandopadhyay (1908-1956).
Seven novels: Padma Nadir Majhi (1936), Dibaratrir Kabya (1936), Chatushkon (1942), Majhir Chhele (1959), and Sahartali, Ahingsa and Darpan, the three collected in the 1965 selection Sera Manik. Novels are split into their printed chapters.
Sixty-one short stories: all fifty-eight of Uttarkaler Galpa-Sangraha (2nd edition, 1964) in their published order, plus… See the full description on the dataset page: https://huggingface.co/datasets/protibimbo/ManikKatha.CC-Bench-trajectories
CC-Bench Trajectories Overview
To evaluate GLM-4.6's agentic coding capabilities in real-world scenarios, we developed CC-Bench-V1.1 using Claude Code as the agentic coding testbed. Building on CC-Bench-V1.0, we added 22 more challenging coding tasks and conducted comprehensive evaluations against Claude-Sonnet-4, GLM-4.5, Kimi-K2-0905, and DeepSeek-V3.1-Terminus. The benchmark comprises 74 coding tasks spanning frontend development, tool development, data analysis, testing, and… See the full description on the dataset page: https://huggingface.co/datasets/ManuelZnnmc/CC-Bench-trajectories.mantra-offset4-lenmatched.k-8.statml-arxiv-qwen3
mantra-offset4-lenmatched.k-8.statml-arxiv-qwen3
A length-matched null control for JackHsieh/luna-reason-only.k-8.statml-arxiv-qwen3.
Every thought is this one sentence, repeated:
We are thinking hard about what comes in the next 8 tokens by reasoning correctly and carefully about what comes before it in the document.
It is fluent, on-topic and completely uninformative — it says nothing about the document it
sits in. Paired against luna at the same cuts, it isolates whether… See the full description on the dataset page: https://huggingface.co/datasets/JackHsieh/mantra-offset4-lenmatched.k-8.statml-arxiv-qwen3.mantra-luna-lenmatched.k-8.statml-arxiv-qwen3.qwen3-ids.kv-tags-explained
mantra-luna-lenmatched.k-8.statml-arxiv-qwen3.qwen3-ids.kv-tags-explained
Tokenized, tag-wrapped form of JackHsieh/mantra-luna-lenmatched.k-8.statml-arxiv-qwen3.
Each thought is wrapped as
<|note|>
This is a hint about a span that appears later in this document. KEY is the text immediately before that span; VALUE is a note about what might come next.
KEY: <last 8 prefix tokens>
VALUE: <thought>
<|/note|>
and stored both as text (thought_text) and as
Qwen/Qwen3-4B-Base token… See the full description on the dataset page: https://huggingface.co/datasets/JackHsieh/mantra-luna-lenmatched.k-8.statml-arxiv-qwen3.qwen3-ids.kv-tags-explained.manipulation-resistant-prompts-1536-1536
Dataset Card: manipulation-resistant-prompts-1536-1536
Dataset Description
This dataset contains prompts with specified target word counts for both input prompts and target outputs, designed to test and evaluate language models across different length requirements. Word counts are defined as whitespace-separated tokens, providing a consistent and human-interpretable measure of text length.
These datasets are typically used in performance benchmarking of language models… See the full description on the dataset page: https://huggingface.co/datasets/metrum-ai/manipulation-resistant-prompts-1536-1536.mantra-luna-lenmatched.k-8.statml-arxiv-qwen3
mantra-luna-lenmatched.k-8.statml-arxiv-qwen3
A length-matched null control for JackHsieh/luna-reason-only.k-8.statml-arxiv-qwen3.
Every thought is this one sentence, repeated:
We are thinking hard about what comes in the next 8 tokens by reasoning correctly and carefully about what comes before it in the document.
It is fluent, on-topic and completely uninformative — it says nothing about the document it
sits in. Paired against luna at the same cuts, it isolates whether the… See the full description on the dataset page: https://huggingface.co/datasets/JackHsieh/mantra-luna-lenmatched.k-8.statml-arxiv-qwen3.manipulation-resistant-prompts-1536-96
Dataset Card: manipulation-resistant-prompts-1536-96
Dataset Description
This dataset contains prompts with specified target word counts for both input prompts and target outputs, designed to test and evaluate language models across different length requirements. Word counts are defined as whitespace-separated tokens, providing a consistent and human-interpretable measure of text length.
These datasets are typically used in performance benchmarking of language models… See the full description on the dataset page: https://huggingface.co/datasets/metrum-ai/manipulation-resistant-prompts-1536-96.mantra-offset4-lenmatched.k-8.statml-arxiv-qwen3.qwen3-ids.kv-tags-explained
mantra-offset4-lenmatched.k-8.statml-arxiv-qwen3.qwen3-ids.kv-tags-explained
Tokenized, tag-wrapped form of JackHsieh/mantra-offset4-lenmatched.k-8.statml-arxiv-qwen3.
Each thought is wrapped as
<|note|>
This is a hint about a span that appears later in this document. KEY is the text immediately before that span; VALUE is a note about what might come next.
KEY: <last 8 prefix tokens>
VALUE: <thought>
<|/note|>
and stored both as text (thought_text) and as
Qwen/Qwen3-4B-Base… See the full description on the dataset page: https://huggingface.co/datasets/JackHsieh/mantra-offset4-lenmatched.k-8.statml-arxiv-qwen3.qwen3-ids.kv-tags-explained.manipulation-resistant-prompts-96-96
Dataset Card: manipulation-resistant-prompts-96-96
Dataset Description
This dataset contains prompts with specified target word counts for both input prompts and target outputs, designed to test and evaluate language models across different length requirements. Word counts are defined as whitespace-separated tokens, providing a consistent and human-interpretable measure of text length.
These datasets are typically used in performance benchmarking of language models, where… See the full description on the dataset page: https://huggingface.co/datasets/metrum-ai/manipulation-resistant-prompts-96-96.Change_Management_1
Change Management 1
This corpus was automatically generated by the Deku Corpus Builder for use in RAG-based AI applications.
Dataset Structure
Each record contains:
text: The content text
source_url: Original source URL
source_title: Title of the source document
source_domain: Domain of the source
license_type: License classification (e.g. public_domain, cc_by, cc_by_sa)
attribution_required: Boolean — True for CC BY / CC BY-SA and other attribution-required licenses… See the full description on the dataset page: https://huggingface.co/datasets/PhillyMac/Change_Management_1.manipulation-resistant-prompts-96-1536
Dataset Card: manipulation-resistant-prompts-96-1536
Dataset Description
This dataset contains prompts with specified target word counts for both input prompts and target outputs, designed to test and evaluate language models across different length requirements. Word counts are defined as whitespace-separated tokens, providing a consistent and human-interpretable measure of text length.
These datasets are typically used in performance benchmarking of language models… See the full description on the dataset page: https://huggingface.co/datasets/metrum-ai/manipulation-resistant-prompts-96-1536.Performance_Management_Difficult_Conversations_Practical
Performance Management Difficult Conversations — Practical
This corpus was automatically generated by the Deku Corpus Builder for use in RAG-based AI applications.
Dataset Structure
Each record contains:
text: The content text
source_url: Original source URL
source_title: Title of the source document
source_domain: Domain of the source
license_type: License classification (e.g. public_domain, cc_by, cc_by_sa)
attribution_required: Boolean — True for CC BY / CC BY-SA and… See the full description on the dataset page: https://huggingface.co/datasets/PhillyMac/Performance_Management_Difficult_Conversations_Practical.Malaysia-Personas
Malaysia-Personas
1,867 synthetic personas of Malaysian voters, in the style of NVIDIA Nemotron-Personas. Each persona is anchored to one real, anonymised record from the published GE15 (2022) electoral roll and fleshed out by an LLM, with calibration to district income and poverty statistics.
They were built for survey simulation: asking a representative synthetic population how it would react to a policy or event, then aggregating the answers.
How it was made… See the full description on the dataset page: https://huggingface.co/datasets/manfye/Malaysia-Personas.sinhala-poems-v1
Sinhala Poems (Filtered)
Curated Sinhala poem blocks extracted from web blogs using a verse-shaped heuristic (v3.4) with weak attributes (theme/mood/style) and stats.
Columns
text: full original block (cleaned)
snippet: first stanza or 8 lines
url: source URL
subtype: poem | song_like | promo_like | unknown
keep: boolean accepted by filter
theme, mood, style, length_class: weak labels
ps_*: structure stats (floats)
Filtering summary
Boilerplate/HTML… See the full description on the dataset page: https://huggingface.co/datasets/manthilaffs/sinhala-poems-v1.Performance_Management_Difficult_Conversations_Theory
Performance Management Difficult Conversations — Theory
This corpus was automatically generated by the Deku Corpus Builder for use in RAG-based AI applications.
Dataset Structure
Each record contains:
text: The content text
source_url: Original source URL
source_title: Title of the source document
source_domain: Domain of the source
license_type: License classification (e.g. public_domain, cc_by, cc_by_sa)
attribution_required: Boolean — True for CC BY / CC BY-SA and… See the full description on the dataset page: https://huggingface.co/datasets/PhillyMac/Performance_Management_Difficult_Conversations_Theory.mantra-14b-user-interaction-log
🧠 Mantra-14B User Interaction Logs
This dataset captures real user interactions with a Gradio demo powered by large-traversaal/Mantra-14B. Each entry logs the user's prompt, the model's response, and additional metadata such as response time and generation parameters. This dataset is ideal for understanding how people engage with the model, evaluating responses, or fine-tuning on real-world usage data.
🔍 What’s Inside
Each row in the dataset includes:
timestamp –… See the full description on the dataset page: https://huggingface.co/datasets/large-traversaal/mantra-14b-user-interaction-log.ManyIH-Bench
ManyIH-Bench
📄 Paper | 🏠 Homepage | 💻 Source Code
ManyIH-Bench is a benchmark for evaluating how well language models resolve instruction conflicts among instructions with arbitrarily many privilege levels. Each sample contains multiple instructions annotated with privilege markers; when instructions conflict, the model should follow the highest-privilege one.
The dataset has two subsets:
coding (427 samples): Code generation tasks (based on MBPP) with conflicting style… See the full description on the dataset page: https://huggingface.co/datasets/jackzhang/ManyIH-Bench.Prioritization_Time_Attention_Management_Practical
Prioritization Time Attention Management — Practical
This corpus was automatically generated by the Deku Corpus Builder for use in RAG-based AI applications.
Dataset Structure
Each record contains:
text: The content text
source_url: Original source URL
source_title: Title of the source document
source_domain: Domain of the source
license_type: License classification (e.g. public_domain, cc_by, cc_by_sa)
attribution_required: Boolean — True for CC BY / CC BY-SA and other… See the full description on the dataset page: https://huggingface.co/datasets/PhillyMac/Prioritization_Time_Attention_Management_Practical.thomas-mann-works
Thomas Mann Works Corpus
This corpus contains 76 German-language works by Thomas Mann: 12 novels, 8 novellas, 19 stories, 27 essays, 8 speeches, 1 play and 1 verse work. Together they contain 2,560,236 words.
Thomas Mann's original German works entered the public domain on January 1, 2026, seventy years after his death in 1955.
Each JSONL record represents one complete work. Collections were separated into their constituent works before export. The corpus is intentionally… See the full description on the dataset page: https://huggingface.co/datasets/phiba/thomas-mann-works.Pomodoro_Technique_for_Time_Management_Corpus
Pomodoro Technique for Time Management
This corpus was automatically generated by the Deku Corpus Builder for use in RAG-based AI applications.
Dataset Structure
Each record contains:
text: The content text
source_url: Original source URL
source_title: Title of the source document
source_domain: Domain of the source
relevance_score: Relevance to the subject (0-1)
quality_score: Content quality score (0-1)
topics: JSON array of detected topics
character_count: Length of… See the full description on the dataset page: https://huggingface.co/datasets/PhillyMac/Pomodoro_Technique_for_Time_Management_Corpus.