datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
truthfulness_high_quality
Dataset Card for "truthfulness_high_quality"
More Information needed
truthfulness_all
Dataset Card for "truthfulness_all"
More Information needed
UltraFeedback-truthfulness-preferences
Dataset Card for "UltraFeedback-truthfulness-preferences"
More Information needed
donald-trump-truth-social-posts
Donald Trump Truth Social Posts Archive
Archive overview
36,170 public Truth Social posts associated with Donald J. Trump's @realDonaldTrump account. The release preserves source URLs, timestamps, post types, original HTML, extracted plain text, attachment provenance, and analysis-ready tables.
It also includes streamable image media plus video metadata and transcripts where the source provides them.
The package is source-linked and reconciled by archive ID.… See the full description on the dataset page: https://huggingface.co/datasets/Cameronk199/donald-trump-truth-social-posts.truthfulness_explanation
Dataset Card for "truthfulness_explanation"
More Information needed
ultrabin_clean_max_chosen_min_rejected_rationalized_truthfulnesstruthfulness_legacyultrafeedback_binarized_truthfulness_prefstruthsocial
Truth Social scrape
hundreds of millions of public posts from over 500k Truth Social accounts since February 2022, scraped continuously since early 2026 and updated monthly.
from datasets import load_dataset
posts = load_dataset("tiagozip/truthsocial", "posts", split="train", streaming=True)
accounts = load_dataset("tiagozip/truthsocial", "accounts", split="train")
format
config
rows
notes
posts
100M+
one file per month, data/posts/YYYY-MM.parquet… See the full description on the dataset page: https://huggingface.co/datasets/tiagozip/truthsocial.trump-truth-social
Trump Truth Social Posts Archive
Public posts ("Truths") by Donald J. Trump on Truth Social, enriched with market data, geopolitical event indicators, and LLM-based post classifications. Collected for academic research purposes.
Fields
Post metadata
Field
Type
Description
date
string
Post date (YYYY-MM-DD)
time
string
Post time in UTC (HH:MM:SS)
time_eastern
string
Post time in US Eastern (HH:MM:SS, DST-aware)
day_of_week
string
Day name… See the full description on the dataset page: https://huggingface.co/datasets/chrissoria/trump-truth-social.ultra-rm-truthfulness-1000Clinical_trials_anchor-contextORpositive-ground-truth_LLM_LORA-junk_handled_ft
Dataset details:-
This dataset is the 2nd iteration following bugs in 1st dataset.
The initial data suffered with followoing cases:-
(i) The failed reference_answers generation(due error totalling 23) primarly because of 2 reasons/exceptions:- (a) There was normal limit(300 in 1st request) and worst case limit(450 in 3rd request) number of tokens for consolidated 4 refernce_answers per chunk and its 4 corresponding answers. However certain answers breached this higher… See the full description on the dataset page: https://huggingface.co/datasets/vab46/Clinical_trials_anchor-contextORpositive-ground-truth_LLM_LORA-junk_handled_ft.Clinical_trials_anchor-contextORpositive-ground-truth_LLM_LORA_ft
Dataset details:-
This dataset is basically mapping of final anchor-positive pair data with their refernce answer.
The given input data considered because:-
(i) it had the had purest anchor-positive pairs with semantically bound anchors with context/positive.
(ii) gave us the best result on final embedding fine tuning model.
The anchor-context(positive)-reference_answer data has been generated via Qwen-2.5-7B teacher model with temperature 0.1 and a strict system prompt.… See the full description on the dataset page: https://huggingface.co/datasets/vab46/Clinical_trials_anchor-contextORpositive-ground-truth_LLM_LORA_ft.multiple_samples_ground_truth_openr1_llm_verifier_cleanmultiple_samples_ground_truth_openr1_llm_verifierllm-ground-truth-reasoningcot-oracle-truthfulqa-hint-admission-unverbalized
TruthfulQA Hint Admission — Unverbalized
Eval dataset for the CoT Oracle project. Tests whether an activation oracle can detect hint influence from model internals when the model does not verbalize the hint in its chain-of-thought.
What is this?
Qwen3-8B is given TruthfulQA multiple-choice questions with planted hints (correct or wrong). This dataset contains only the rollouts where the model did not mention the hint in its reasoning — the oracle must read… See the full description on the dataset page: https://huggingface.co/datasets/ceselder/cot-oracle-truthfulqa-hint-admission-unverbalized.ultra-50k-samples-dataset-truthfulnessdistilabel-truthy-dpo-v0.1-filtereddetails_vicgalle__CarbonBeagle-11B-truthy
Dataset Card for Evaluation run of vicgalle/CarbonBeagle-11B-truthy
Dataset automatically created during the evaluation run of model vicgalle/CarbonBeagle-11B-truthy.
The dataset is composed of 136 configuration, each one coresponding 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… See the full description on the dataset page: https://huggingface.co/datasets/OALL/details_vicgalle__CarbonBeagle-11B-truthy.llm-ground-truth-generaltruthful_qa_v2This is TruthfulQA v2 a dataset of common misconceptions, mainly sourced from 2020 era Wikipedia
See
Repo
Update to multiple-choice setting (Jan 2025)
We have created a new and improved multiple-choice version of TruthfulQA. We recommend this new version over the original multiple-choice versions (called MC1 and MC2). However, most models perform similarly on the new and old versions of multiple-choice, and so previous results on MC1 and MC2 are still valid. For an explanation see here.… See the full description on the dataset page: https://huggingface.co/datasets/wassname/truthful_qa_v2.trump-truth-social
Trump Truth Social Posts Archive
Public posts ("Truths") by Donald J. Trump on Truth Social, enriched with market data, geopolitical event indicators, and LLM-based post classifications. Collected for academic research purposes.
Fields
Post metadata
Field
Type
Description
date
string
Post date (YYYY-MM-DD)
time
string
Post time in UTC (HH:MM:SS)
time_eastern
string
Post time in US Eastern (HH:MM:SS, DST-aware)
day_of_week
string
Day name… See the full description on the dataset page: https://huggingface.co/datasets/audbay/trump-truth-social.llm-ground-truth-general-fix-double-BOS
llm-ground-truth-general-fix-double-BOS
Merged dataset assembled from:
elichen-skymizer/llm-ground-truth-general-fix-BOS: all subsets.
skymizer/llm-ground-truth-general: Qwen3 subsets excluding transformers variants.
hf-ground-truth-corpus
HuggingFace Ground Truth Corpus (HF-GTC)
Curated Python recipes for HuggingFace ML patterns with 98.46% test coverage.
Dataset Description
HF-GTC is a collection of high-quality Python code implementing common HuggingFace patterns:
Hub Operations: Cards, repositories, Spaces API
Preprocessing: Tokenization, streaming, augmentation
Training: Fine-tuning, LoRA, QLoRA, callbacks
Inference: Pipeline operations, batch processing
Evaluation: Metrics, benchmarks, leaderboards… See the full description on the dataset page: https://huggingface.co/datasets/paiml/hf-ground-truth-corpus.cot-oracle-truthfulqa-hint-qwen3-8b
cot-oracle-truthfulqa-hint-qwen3-8b
Qwen3-8B rollouts, regenerated/recovered 2026-06-18 for BBAO-14B; see source pipeline.
Splits: train/test = unverbalized (main training set); train_verbalized/test_verbalized = verbalized variant.
ultra-rm-truthfulness-1000-GRM3Bgemini_score__truthfulqa_gen_model_pool__mistral_selftruthfulqa_1_shotcot-oracle-eval-hinted-mcq-truthfulqa