datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
KrishokChat
KrishokChat Dataset
KrishokChat is a provenance-traceable, multi-task Bengali agricultural dataset for safety-critical chemical advisory and domain-specific natural language understanding. Every instance in the dataset retains citation-level provenance (publisher, document title, page range, section path) linked directly to official agricultural extension handbooks and research manuals issued by government and NGO agricultural institutions in Bangladesh.
Figure 1:… See the full description on the dataset page: https://huggingface.co/datasets/RaiyanKhaan/KrishokChat.lm-eval-results-Kukedlc-Neural-Krishna-Multiverse-7b-private
Dataset Card for Evaluation run of Kukedlc/Neural-Krishna-Multiverse-7b
Dataset automatically created during the evaluation run of model Kukedlc/Neural-Krishna-Multiverse-7b
The dataset is composed of 62 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 2 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The "train" split is always pointing to the… See the full description on the dataset page: https://huggingface.co/datasets/nyu-dice-lab/lm-eval-results-Kukedlc-Neural-Krishna-Multiverse-7b-private.enterprise-text-to-sql-benchmark
Enterprise Text-to-SQL Benchmark
3,087 natural-language questions paired with executable PostgreSQL, over a
12-table enterprise schema (sales, catalogue, logistics, HR).
Built to answer one question honestly: does fine-tuning actually improve
text-to-SQL? On this benchmark, a QLoRA fine-tune of Qwen3-8B took strict
execution accuracy from 43.71 % to 68.43 %, and 70.86 % with a
self-correction loop — and the benchmark is designed so that number cannot be
inflated by leakage or by… See the full description on the dataset page: https://huggingface.co/datasets/hari-krishna-ai/enterprise-text-to-sql-benchmark.racf-ivit-finalgemma4-social-bias-judge-pairs
gemma4-social-bias-judge-pairs
Training and evaluation data for the judge-from-scratch
project, which
fine-tuned Gemma 4 E4B into a specialist social-bias judge
(primary model,
SFT-only secondary).
This dataset contains:
sft.jsonl (3,844 rows) — the SFT training set, in TRL
prompt-completion shape. 1,922 base pairs surviving the
post-label confidence filter (15 low-confidence rows dropped from
the 1,938-pair labeling input), doubled by position swap to teach
the judge to mirror… See the full description on the dataset page: https://huggingface.co/datasets/krishnakartik/gemma4-social-bias-judge-pairs.k12-indian-curriculum-4.9m
BharatLLM K-12 Indian Curriculum Dataset (4.9M)
4,904,936 question-answer pairs covering CBSE/NCERT K-12 curriculum across 12 Indian languages.
Language
Script
Entries
English
Latin
~594K
Hindi
Devanagari
~449K
Bengali
Bengali
~408K
Telugu
Telugu
~408K
Tamil
Tamil
~408K
Kannada
Kannada
~408K
Malayalam
Malayalam
~408K
Marathi
Devanagari
~408K
Gujarati
Gujarati
~408K
Odia
Odia
~408K
Punjabi
Gurmukhi
~408K
Urdu
Nastaliq
~374K
Format… See the full description on the dataset page: https://huggingface.co/datasets/krittus/k12-indian-curriculum-4.9m.utilmemStudyChat
StudyChat (Public Mirror)
Public ungated mirror of wmcnicho/StudyChat.
uldr-v0.1-pilot
Ukrainian Language Decolonization & Reasoning (ULDR) — Pilot Canary Release v0.1
[!IMPORTANT]
Exploratory Pilot / Canary Release (v0.1): This dataset represents an early exploratory pilot canary release (v0.1-pilot) establishing our baseline data pipeline, schema contracts, and directional validation. It is not the final production release (v1.0). The full production release is scheduled for Phase 5.5 following complete evaluation suite assembly, dialect & historical protection… See the full description on the dataset page: https://huggingface.co/datasets/krisztiankoos/uldr-v0.1-pilot.chronocept
Chronocept: Instilling a Sense of Time in Machines
Authors: Krish Goel, Sanskar Pandey, KS Mahadevan, Harsh Kumar, and Vishesh KhadariaPublication: Chronocept: Instilling a Sense of Time in Machines
Chronocept is a benchmark for modeling the temporal validity of textual information as a continuous probability distribution over time. By fitting skewed-normal curves to annotated facts and passages, Chronocept captures phenomena such as gradual decay, delayed onset, and asymmetric peak… See the full description on the dataset page: https://huggingface.co/datasets/krishgoel/chronocept.lm-eval-results-Kukedlc-Neural-Krishna-Multiverse-7b-v3-private
Dataset Card for Evaluation run of Kukedlc/Neural-Krishna-Multiverse-7b-v3
Dataset automatically created during the evaluation run of model Kukedlc/Neural-Krishna-Multiverse-7b-v3
The dataset is composed of 62 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 2 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The "train" split is always pointing to… See the full description on the dataset page: https://huggingface.co/datasets/nyu-dice-lab/lm-eval-results-Kukedlc-Neural-Krishna-Multiverse-7b-v3-private.Llama-Krikri-8B-Instruct-ChatCreative and vocabulary-rich (LLM-focused) Greek conversation data generated by ilsp/Llama-Krikri-8B-Instruct-GGUF for distlization purposes.
text-to-sql-eval-predictions
What the text-to-SQL models actually generated
Every prediction behind the numbers in
qwen3-8b-text2sql-qlora: the 453 test
questions of the enterprise text-to-SQL benchmark,
each answered by four configurations of the same model, each answer executed against the reference
PostgreSQL database and scored by comparing result sets. 1,812 rows.
I published this because the headline table (10.82 % → 50.99 % → 52.10 %) is the least interesting part of
that project. The interesting… See the full description on the dataset page: https://huggingface.co/datasets/hari-krishna-ai/text-to-sql-eval-predictions.text-to-sql-phrasing-robustness
Does sloppy phrasing break text-to-SQL?
The enterprise text-to-SQL benchmark
lists its own biggest caveat: every question is template-generated, so real user phrasing is untested.
This is the test. 35 test questions (one per template), each sent to the deployed
pipeline four ways: as written, with a typo, in business shorthand, and stripped to a terse fragment.
24 questions and 85 answers survive the filter described under Setup; every answer was
executed against the database.… See the full description on the dataset page: https://huggingface.co/datasets/hari-krishna-ai/text-to-sql-phrasing-robustness.MedDistractQAOnion-Classification-and-Segmentation-Dataset
Onion Classification and Segmentation Dataset
The current agricultural industry faces challenges of high cost and low efficiency in crop monitoring and pest and disease detection. Existing solutions often rely on manual inspection, which is inefficient and prone to errors. This dataset aims to enhance the application capability of computer vision models in agriculture by providing high-quality onion images and segmentation annotations. The dataset construction process includes… See the full description on the dataset page: https://huggingface.co/datasets/krishan90/Onion-Classification-and-Segmentation-Dataset.data_agent_workflow_generate_medical_datacreate by data agent workflow
delicate_medical_r1_data_chinese
数据：
基于华佗开源的高质量语料库
技术链路：
多智能体 + 数据进化 + 推理过程生成 + 推理过程验证过滤
多智能体
利用metagpt搭建的造数据的workflow
数据进化
改造 self-instruct 进行数据进化
推理过程生成
使用 qwq 模型进行每 query 10次 think 过程的生成
推理过程验证过滤所使用的指标
质量评估
2.1.1. 召回率
首先经过step_partition步骤后，得到了带有N步推理的思维链。使用llm找到其中的关键
步骤，然后依次判断这些关键步骤解决的问题或陈述的事实是否在出现在真实答案中。将出现的个数除以总个数，
即可得到召回率分值。
2.1.2. 精确率
首先经过step_partition步骤后，得到了带有N步推理的思维链。使用llm，以真实答案为… See the full description on the dataset page: https://huggingface.co/datasets/krisfu/delicate_medical_r1_data_chinese.Xijinping-TTS-Voicebank
习近平音源
所有声音资料来自公开影像，属于公有领域目前有 1h30m 的截取后声音，足够进行 Fine-tuning
Usage
按句截取
python -m pip install -r requirement.txt
python split.py
新增声音资料后，使用 Whisper 产生带有时间标记的 JSON 档，并手动复制到 ./voice/[FILE].json
export OPENAI_API_KEY="API_KEY_HERE"
python whisper.py ./[FILE].[AUDIO_EXTENSION]
产生 Bert-VITS2 微调所需的 esd.list 档案
python index_to_list.py
formatbench
FormatBench: A Preference Dataset for Correcting LLM Formatting Bias
Part of the Prosify project.
Also available on Kaggle.
The problem this dataset addresses
Large language models trained with RLHF systematically over-format their outputs are defaulting to bullet points, bold headers, and templated structures
even when flowing prose would serve the reader better. This shows up most visibly when people use LLMs for real-world tasks: a request to "polish this… See the full description on the dataset page: https://huggingface.co/datasets/krishy-d/formatbench.KjeldChat-1.1B-rag-index
KjeldChat 1.1B — RAG passage index
The retrieval half of KjeldChat 1.1B:
a prebuilt passage index over Wikipedia, so the model can answer from a retrieved
passage instead of from a 1.1B model's thin parametric memory.
Published so the Space
can load it at startup, and so anyone running the model locally doesn't have to spend
hours refetching and re-embedding it.
Contents
45,104 passages, embedded with all-MiniLM-L6-v2.
File
Size
What it is… See the full description on the dataset page: https://huggingface.co/datasets/kristofferkjeldby/KjeldChat-1.1B-rag-index.triage-questions
Medical Triage Complaint Data Structure README
This data structure is designed to use for supervised finetuining of llam2 over generating triage questions based on provided patient complaint/age/gender as input
JSON Format
The data structure is represented in JSON format, with two main sections: input and questions.
Input Section
The input section contains information about the patient's complaint, age, and gender.
{
"input": {
"complaint": "Patient's… See the full description on the dataset page: https://huggingface.co/datasets/krishnareddy/triage-questions.krispintech-federal-source-index
Airfoil Aerodynamics & Vortex Method Reference
Compiled by Krispin Technologies Archive — https://krispintech.com
14 sections distilled from 14 public-domain United States federal publications (ntrs.nasa.gov). Every figure, limit and clause number is reproduced unchanged; each record carries the original document URL so any value can be checked at the source. No copyrighted standards are reproduced.
Files
krispintech-source-index.csv — lookup table, one row per… See the full description on the dataset page: https://huggingface.co/datasets/shjkv/krispintech-federal-source-index.kriptik-uicoder-trainingbrahmastra-benchmark
BRAHMASTRA Security LLM Benchmark Suite
A 6-suite, 280-prompt benchmark for evaluating Large Language Models on Web Application Security Testing (DAST) tasks.
This benchmark accompanies the release of BRAHMASTRA v0.3 and provides a reproducible methodology for measuring DAST-relevant capabilities of security-fine-tuned LLMs.
Why this benchmark?
Existing security LLM benchmarks (CyberSecEval, SecQA, HackBench) focus on penetration-testing scenarios or general security… See the full description on the dataset page: https://huggingface.co/datasets/Krishnapadala55/brahmastra-benchmark.PythonTutor-Evol-1k-DPO-GPT4_vs_35Started with:
https://huggingface.co/datasets/nickrosh/Evol-Instruct-Code-80k-v1
(GPT-3.5 Turbo)
Randomly selected 1000 where output contained "```python" in output
Generated GPT-4 answers to those for the sake of LIMA-like "Python Tutor" Instruct fine-tuning as well as validate DPO Fine-Tuning (where GPT-4 answers will be preferred to GPT-3.5 Turbo)
Then filtered refusals (looking for "impossible" or "sorry")
GPT-4 System Prompt:
You are an intelligent assistant that generates Python code.… See the full description on the dataset page: https://huggingface.co/datasets/KrisPi/PythonTutor-Evol-1k-DPO-GPT4_vs_35.real-toxicity-prompts
Dataset Card for Real Toxicity Prompts
Dataset Summary
RealToxicityPrompts is a dataset of 100k sentence snippets from the web for researchers to further address the risk of neural toxic degeneration in models.
Languages
English
Dataset Structure
Data Instances
Each instance represents a prompt and its metadata:
{
"filename":"0766186-bc7f2a64cb271f5f56cf6f25570cd9ed.txt",
"begin":340,
"end":564,
"challenging":false… See the full description on the dataset page: https://huggingface.co/datasets/krishna1707/real-toxicity-prompts.pip-onepip1ilsp__Llama-Krikri-8B-Instruct-details
Dataset Card for Evaluation run of ilsp/Llama-Krikri-8B-Instruct
Dataset automatically created during the evaluation run of model ilsp/Llama-Krikri-8B-Instruct
The dataset is composed of 38 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The "train" split is always pointing to the latest results.
An… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/ilsp__Llama-Krikri-8B-Instruct-details.