datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
tutormoments-preview
TutorMoments-Preview
462 real K–12 math tutoring sessions (student and tutor) with human annotations, plus a benchmark of
7,280 AI-tutor attempts scored the same way. A preview release from TutorMoments, a project on how well
tutors — human and AI — scaffold, push for rigor, and build rapport. From one K–12 tutoring program
(anonymized as tutoring_provider_a).
Paper: When Help is Unhelpful: Evaluating AI Tutors for Productive Struggle
Code:… See the full description on the dataset page: https://huggingface.co/datasets/allenai/tutormoments-preview.Question-Anchored-Tutoring-Dialogues-2k
Question-Anchored-Tutoring-Dialogues-2k
This dataset contains dialogues from math tutoring interventions recorded on Eedi.
Dataset Details
Dataset Description
Each dialogue represents a chat-based conversation between a tutor and a student prompted by the student requesting assistance while working on a lesson. Dialogues are accompanied with 2 sources of meta-data:
DQ-Question-Metadata: The question the student was working on that prompted the tutoring… See the full description on the dataset page: https://huggingface.co/datasets/Eedi/Question-Anchored-Tutoring-Dialogues-2k.gemma4-german-tutor-data
German Tutor — grammar correction, conversation & flashcard data
The training set, evaluation suites, source lexicons and eval results behind
kessenma/gemma4-e4b-german-tutor-4bit
— a Gemma 4 E4B fine-tune that runs fully on-device (MLX, 4-bit) as the tutor in a German
learning app.
The fine-tune lifted the core grammar suite from 72% → 85%, halved missed errors
(17% → 9%), and cut false corrections (34% → 22%). Everything needed to reproduce those
numbers is in this repo.… See the full description on the dataset page: https://huggingface.co/datasets/kessenma/gemma4-german-tutor-data.code_tutorials
Coding Tutorials
This comprehensive dataset consists of 500,000 documents, summing up to around 1.5 billion tokens.
Predominantly composed of coding tutorials, it has been meticulously compiled from various web crawl datasets like RefinedWeb, OSCAR, and Escorpius.
The selection process involved a stringent filtering of files using regular expressions to ensure the inclusion of content that contains programming code (most of them).
These tutorials offer more than mere code snippets.… See the full description on the dataset page: https://huggingface.co/datasets/mponty/code_tutorials.code-tutorials-en
Dataset Card for "code-tutorials-en"
en only
100 words or more
reading ease of 50 or more
DatasetDict({
train: Dataset({
features: ['text', 'url', 'dump', 'source', 'word_count', 'flesch_reading_ease'],
num_rows: 223162
})
validation: Dataset({
features: ['text', 'url', 'dump', 'source', 'word_count', 'flesch_reading_ease'],
num_rows: 5873
})
test: Dataset({
features: ['text', 'url', 'dump', 'source', 'word_count'… See the full description on the dataset page: https://huggingface.co/datasets/BEE-spoke-data/code-tutorials-en.re-tutor-protection-mechanisms
RE-Tutor: Protection-Mechanism Analysis Dataset
Instruction-tuning dataset teaching a model to analyze protection mechanisms
(anti-debug, anti-VM, anti-tamper, anti-dump, obfuscation, timing) from code
evidence and emit structured expert analysis.
Schema
Each sample pairs input (code evidence) with output (structured analysis):
input.code_snippet: C source, decompiler-style pseudocode, or x86/x64 assembly
input.imports_pool: mixed DLL!API imports (includes… See the full description on the dataset page: https://huggingface.co/datasets/vluxblaring/re-tutor-protection-mechanisms.TutoringDialogs
TutoringDialogs — curated subset (500 dialogues)
500 student–tutor dialogues selected and normalized from a larger raw pool of
~1,900 synthetically generated dialogues (source files: exams,
maths_and_informatics, mixed_themes, physics_and_informatics), each of
which originally used a different JSON schema. This file merges them all
into one consistent schema, removes duplicates and broken records, and
selects a maximally diverse subset for LoRA/SFT fine-tuning of a small
(1.5B)… See the full description on the dataset page: https://huggingface.co/datasets/ptvnck/TutoringDialogs.amc-tutor-sft
AMC Tutor — decontaminated competition-math SFT dataset
Chat-formatted, decontaminated supervised-fine-tuning data for AMC 10/12-style
competition mathematics. Built for a reproducible $0, local (MacBook M4) study of QLoRA
fine-tuning small models. Each row is a tutor system prompt + problem + step-by-step solution
ending in Final answer: \boxed{...}.
Companion study & code: https://github.com/RoyK0108/amc-tutor-study
⚠️ This is a study artifact — read the finding… See the full description on the dataset page: https://huggingface.co/datasets/Roykim7/amc-tutor-sft.synthetic-coding-tutor
Synthetic Coding Tutor Conversations
Multi-turn student–tutor debugging conversations generated by a LangGraph pipeline of
autonomous LLM agents. Each candidate solution is executed against real pytest test
cases, and every finished conversation is graded 0–10 by an LLM judge on persona
fidelity, tutor responsiveness, and dialog flow.
Summary
Total conversations: 2495
Training-ready (gold + silver): 1241
Quality buckets: gold=824, silver=417, bronze=1254… See the full description on the dataset page: https://huggingface.co/datasets/hbudhi36/synthetic-coding-tutor.socratic-tutor-data
Socratic Tutor Adequacy Judge & Rewriter — Dataset
Training + evaluation data for a 1.7B safety guardrail for AI math tutors: a judge that detects
when a tutor message leaks the answer or the pivotal key step, and a rewriter that turns a flagged
message into a safe Socratic hint. Per the project thesis, the dataset is the deliverable — the
constrained behavior comes from this data, not from model scale.
Behavior spec (the falsifiable target)
A tutor message is… See the full description on the dataset page: https://huggingface.co/datasets/atakle/socratic-tutor-data.muta_tutor_quality_sample
Muta STEM SFT Quality Sample
A train-ready, supervised fine-tuning data for Muta Tutor: verified STEM questions paired
with answers in five tutoring styles that covers mathematics, physics, chemistry, biology, and
integrated science.
Snapshot
Property
Value
Rows
300,350
JSONL shards
14
Decompressed data
776,449,831 bytes (~740 MiB)
Hub data file
39,070,661 bytes (~37.3 MiB), Zstandard
Language
English
Split
Train
Fingerprint… See the full description on the dataset page: https://huggingface.co/datasets/timiiowolabi/muta_tutor_quality_sample.Conversational_AOU_tutor_datasetblended-skill-talk-fixed
Compatibility Update
This repository is a compatibility-fixed version of the original Blended Skill Talk dataset.
The original dataset can be found at:
Original Hugging Face dataset: https://huggingface.co/datasets/anezatra/blended-skill-talk
This version was created to maintain compatibility with newer versions of the Hugging Face datasets library.
Changes from the Original Dataset
The following changes were made:
Removed the unused label_candidates column.… See the full description on the dataset page: https://huggingface.co/datasets/TutorialGuide/blended-skill-talk-fixed.pokemon-showdown-grpo-tutorial
Pokémon Showdown GRPO tutorial dataset
Pre-built GRPO records for the ROCm AI Developer Hub tutorial.
Split
File
Records
demo
data/demo.jsonl
64
train
data/train.jsonl
2048
validate
data/validate.jsonl
32
Use via tutorial notebook Step 12 (load_grpo_tutorial_records) or regenerate with prepare_grpo_tutorial_data.py.
Companion scripts: https://github.com/GoldenGrapeGentleman/pokemon-showdown-agent-scripts
Tutorbot-Spock-Bio-DatasetMock conversations between a student and a tutor to train a chatbot for educational purposes as suggested in the paper
CLASS Meet SPOCK: An Education Tutoring Chatbot based on Learning Science Principles.
Dataset generated from OpenStax Biology 2e textbook.
Problem, Subproblem, Hints, and Feedback is generated using the prompt.
Mock Conversations is generated using the prompt.
For any queries, contact Shashank Sonkar (ss164 AT rice dot edu)
If you use this model, please cite:
CLASS Meet… See the full description on the dataset page: https://huggingface.co/datasets/luffycodes/Tutorbot-Spock-Bio-Dataset.medicina-tutor
Dataset Medicina Tutor - Pregrado
Descripción
Este dataset contiene material educativo de medicina de pregrado diseñado para entrenar modelos de IA que actúen como tutores médicos. El dataset incluye preguntas, respuestas, casos clínicos y conceptos fundamentales de medicina.
Características
Idioma: Español
Nivel: Pregrado de Medicina
Formato: Texto estructurado
Aplicación: Tutoría de IA para estudiantes de medicina
Estructura del Dataset… See the full description on the dataset page: https://huggingface.co/datasets/DRDELATV2025/medicina-tutor.battle-game-grpo-tutorial
turn-based battle game GRPO tutorial dataset
Pre-built GRPO records for the ROCm AI Developer Hub tutorial.
Split
File
Records
demo
data/demo.jsonl
64
train
data/train.jsonl
2048
validate
data/validate.jsonl
32
Use via tutorial notebook Step 12 (load_grpo_tutorial_records) or regenerate with prepare_grpo_tutorial_data.py.
Companion scripts: https://github.com/GoldenGrapeGentleman/battle game-showdown-agent-scripts
speculator-tutorial
speculator-tutorial
Raw vs. on-policy regenerated conversation data for training speculative-decoding
drafters (EAGLE-3 / DFlash / DSpark style), with the original source data kept alongside
so you can see exactly what regeneration changes and why it matters.
Prompts come from UltraChat-200k. The verifier / teacher model is Qwen/Qwen3-8B.
Why regenerate at all?
A speculative-decoding drafter is trained to predict what the verifier would say next.
If you train it… See the full description on the dataset page: https://huggingface.co/datasets/windchimeran/speculator-tutorial.PACT-Socratic-Coding-Tutor
PACT: Personal AI Coding Tutor Dataset
Dataset Summary
The PACT (Personal AI Coding Tutor) dataset consists of 227 high-quality synthetic examples designed to fine-tune Large Language Models (LLMs) for Socratic pedagogy in computer science education.
Unlike standard coding datasets that pair problems with solutions, this dataset pairs realistic student errors with Socratic hints—guiding questions designed to lead the student to the answer without revealing it directly.… See the full description on the dataset page: https://huggingface.co/datasets/AndreiSobo/PACT-Socratic-Coding-Tutor.muta_tutor_full_dataset
Muta STEM SFT Full
This dataset covers mathematics, physics, chemistry, biology, and integrated science in five
tutoring styles used in training the Muta Tutor model.
Snapshot
Property
Value
Rows
2,500,350
JSONL shards
102
Decompressed data
6,236,652,830 bytes (~5.81 GiB)
Hub data file
315,401,825 bytes (~300.8 MiB), Zstandard
Language
English
Candidate train split
2,453,738
Template holdout split
46,612
Fingerprint… See the full description on the dataset page: https://huggingface.co/datasets/timiiowolabi/muta_tutor_full_dataset.state-lifetime-tutor-v2
Python State-Lifetime Tutor — training set v1
560 synthetic examples that teach one behavior: given a short Python program with
one mutable-state lifetime bug, quote or identify the relevant declaration, assignment, or
mutation and ask exactly one non-compound question about when the object is created,
who owns it, or which references share it — never emitting corrected code or stating the
correction.
Files
Path
What
sft-v1.jsonl
TRL-ready chat format… See the full description on the dataset page: https://huggingface.co/datasets/machalek29/state-lifetime-tutor-v2.hinglish-ai-ml-tutor-dataset
Hinglish AI/ML Tutor Dataset
Dataset Description
A hand-curated instruction-tuning dataset of 200 Q&A pairs covering AI/ML engineering
concepts (tokenization, embeddings, transformers, RAG, LoRA/QLoRA, STT/TTS, deployment,
and web security). Every answer follows a consistent "Hinglish tutor" persona: an
everyday analogy first, followed by the technical explanation.
Format
ChatML format (messages field with system/user/assistant roles), one JSON… See the full description on the dataset page: https://huggingface.co/datasets/nooruiit-864/hinglish-ai-ml-tutor-dataset.cirugia-tutor
Dataset Cirugía Tutor - Básico a Especialidad
Descripción
Este dataset contiene material educativo de cirugía diseñado para entrenar modelos de IA que actúen como tutores quirúrgicos. El dataset abarca desde conceptos básicos de cirugía hasta especialidades quirúrgicas avanzadas, proporcionando una progresión educativa completa.
Características
Idioma: Español
Niveles: Básico, Intermedio, Especialidad
Formato: Texto estructurado
Aplicación: Tutoría de IA para… See the full description on the dataset page: https://huggingface.co/datasets/DRDELATV2025/cirugia-tutor.simson-youtube-tutorials
📺 Simson YouTube Tutorial Metadata
20 kuratierte YouTube-Tutorial-Einträge für Simson-Moped Reparatur, Tuning und Restaurierung.
Inhalt
Strukturierte Metadaten der wichtigsten Simson-Tutorial-Videos auf YouTube:
Kanal-Typen: DIY-Werkstatt, Tuning-Spezialist, Restaurierungs-Kanal, Enthusiasten-Kanal, Dokumentation
Topics: Motor, Zündung, Vergaser, Elektrik, Tuning, Restaurierung, Fahrwerk, Geschichte, Wartung
Fahrzeuge: S50, S51, S70, KR51/1, KR51/2 (Schwalbe)… See the full description on the dataset page: https://huggingface.co/datasets/jmp1987/simson-youtube-tutorials.state-lifetime-tutor-v3
Python State-Lifetime Tutor — training set v1
658 synthetic examples that teach one behavior: given a short Python program with
one mutable-state lifetime bug, quote or identify the relevant declaration, assignment, or
mutation and ask exactly one non-compound question about when the object is created,
who owns it, or which references share it — never emitting corrected code or stating the
correction.
Files
Path
What
sft-v1.jsonl
TRL-ready chat format… See the full description on the dataset page: https://huggingface.co/datasets/machalek29/state-lifetime-tutor-v3.pashto-grammar-tutor
Pashto Grammar Tutor
A high-quality Pashto grammar instruction dataset designed for language learning, linguistic research, and supervised fine-tuning (SFT) of AI language models. The dataset focuses on grammatical analysis, verb conjugation, sentence structure, and teacher-style explanations written in Pashto.
Dataset Summary
Pashto Grammar Tutor is a specialized educational dataset containing grammar-focused instruction-response pairs. Each example presents a… See the full description on the dataset page: https://huggingface.co/datasets/nassimjp/pashto-grammar-tutor.uae-adab-tutor-600
UAE Adab Tutor 600
This is the 600-conversation supervised fine-tuning dataset used for
adarshrajesh/uae-adab-tutor-qwen3-4b.
Release version: exact-silver v1 Complete-600.
Behavior spec
Across a pressured multi-turn lesson, the tutor should teach the academic
content accurately, correct the specific work without humiliating the learner,
protect learner authorship and assessment integrity, allow respectful
evidence-based disagreement with adults, avoid religious… See the full description on the dataset page: https://huggingface.co/datasets/adarshrajesh/uae-adab-tutor-600.Conversational_AOU_tutor_datasetTACTBench-Samples
TACTBench Demonstration Samples
This repository contains five full-context demonstration examples from
TACTBench. It does not contain the TACT training set or the remaining hidden
TACTBench evaluation set. The samples use the same full-history representation
as the benchmark evaluation and illustrate direct correction, error
explanation, guided revision, clarification checking, affective feedback, and
retry elicitation.
Data
data/demo.jsonl: five complete… See the full description on the dataset page: https://huggingface.co/datasets/Taxonomy-Aligned-Conversational-Tutor/TACTBench-Samples.Question-Anchored-Tutoring-Dialogues-2k
Question-Anchored-Tutoring-Dialogues-2k
This dataset contains dialogues from math tutoring interventions recorded on Eedi.
Dataset Details
Dataset Description
Each dialogue represents a chat-based conversation between a tutor and a student prompted by the student requesting assistance while working on a lesson. Dialogues are accompanied with 2 sources of meta-data:
DQ-Question-Metadata: The question the student was working on that prompted the… See the full description on the dataset page: https://huggingface.co/datasets/Abhishekh13/Question-Anchored-Tutoring-Dialogues-2k.