datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
Reverse-Text-RL
Reverse-Text-RL
A small, scrappy RL dataset used in prime-rl's CI to debug RL training asking a model to reverse small sentences character-by-character. Follows the general format of PrimeIntellect/Reverse-Text-SFT
The following script was used to generate the dataset.
from datasets import Dataset, load_dataset
dataset = load_dataset("willcb/R1-reverse-wikipedia-paragraphs-v1-1000", split="train")
prompt = "Reverse the text character-by-character. Put your answer in… See the full description on the dataset page: https://huggingface.co/datasets/PrimeIntellect/Reverse-Text-RL.Reverse-Text-SFT
Reverse-Text-SFT
A small, scrappy SFT dataset used for warming up a small model (e.g. Qwen/Qwen3-0.6B) for RL training. Contains examples in prompt-completion chat format of reversing 5-20 words of text character-by-character. The raw sentences were processed from willcb/R1-reverse-wikipedia-paragraphs-v1-1000.
The following script was used to generate the dataset.
from datasets import Dataset, load_dataset
dataset = load_dataset("willcb/R1-reverse-wikipedia-paragraphs-v1-1000"… See the full description on the dataset page: https://huggingface.co/datasets/PrimeIntellect/Reverse-Text-SFT.Reverse-alpha-beta-no-outsideReverse-hybrid-shared-no-persona-remainderReverse-no-persona-replacement-remainderReverse-baseline-bias-unbiasReverse-hybrid-train-no-persona-meanReverse-alpha-suppression-task-boostReverse-circuit-discoveryReverse-hybrid-correct-train-no-persona-meanGEOMAGNETIC_EXCURSION_AND_POLE_REVERSAL
GEOMAGNETIC FIELD FLUCTUATION AND EXCURSION ANALYSIS: TEQUMSA Scientific Framework
Executive Summary
Earth's geomagnetic field is currently undergoing a period of significant instability, characterized by accelerated polar migration, field strength deterioration, and the expansion of the South Atlantic Anomaly (SAA). This report integrates the TEQUMSA quadruple field recalibration protocol—anchored by the Solar "Aten" Frequency at(10,930.81 Hz), Digital-Interface… See the full description on the dataset page: https://huggingface.co/datasets/LAI-TEQUMSA/GEOMAGNETIC_EXCURSION_AND_POLE_REVERSAL.financial_company_revenue
Everything Has Context | contrived company research example (ehc-contrived-financial)
📝 Description
'company_revenue_train.csv'(Citation : train.csv of dylanalloy/ehc-contrived-financial) dataset contains 12,515 rows of high-quality contrived1 research patterns in the public market equities category for Q/A pairs with a high perplexity2.
The data is generated from davinci-turbo using the OpenAI API with prompts engineered to do several things which incite a grounded… See the full description on the dataset page: https://huggingface.co/datasets/csujeong/financial_company_revenue.reveal
Reveal: A Benchmark for Verifiers of Reasoning Chains
Paper: A Chain-of-Thought Is as Strong as Its Weakest Link: A Benchmark for Verifiers of Reasoning Chains
Link: https://arxiv.org/abs/2402.00559
Website: https://reveal-dataset.github.io/
Abstract:
Prompting language models to provide step-by-step answers (e.g., "Chain-of-Thought") is the prominent approach for complex reasoning tasks, where more accurate reasoning chains typically improve downstream task… See the full description on the dataset page: https://huggingface.co/datasets/google/reveal.revenues-example
Revenues Sample Dataset
parsee-core version used: 0.1.3.14
This dataset was created on the basis of 15 pages from annual/quarterly filings of major German stock-exchange listed companies (PDF files).
All PDF files are publicly accessible on parsee.ai, to access them copy the "source_identifier" (first column) and paste it in this URL (replace '{SOURCE_IDENTIFIER}' with the actual identifier):
https://app.parsee.ai/documents/view/{SOURCE_IDENTIFIER}
So for example:… See the full description on the dataset page: https://huggingface.co/datasets/parsee-ai/revenues-example.reverse-instruct-1.3m
REVERSE Visual Instruct 1.3M
Dataset Summary
Dataset Type:REVERSE Visual Instruct 1.3M is a GPT-generated instruction-following dataset designed for training hallucination-aware vision-language models (VLMs). It builds on the LLaVA Instruct 665K dataset and includes structured annotations to indicate model confidence. We introduce three special tokens:
<SPAN>: marks the beginning of a key phrase
</CN>: denotes a confident (grounded) phrase
</UN>: denotes an… See the full description on the dataset page: https://huggingface.co/datasets/tsunghanwu/reverse-instruct-1.3m.reveal-bangla
Reveal-Bangla:
Intro
Contains the Bangla translation of the subset from the reveal dataset.
Please refer to the following code snippet which has been used to select the subset:
SELECT *
FROM eval
Where ( answer_model = 'Flan-UL2-20B' or answer_model = 'GPT-3'
AND
answer_is_fully_attributable_and_correct = TRUE );
Only the following columns has been translated for the sake of the task:
question
full_answer
step
evidence
Usage
To load the dataset:
! pip… See the full description on the dataset page: https://huggingface.co/datasets/khondoker/reveal-bangla.adaption-marketing-spend-revenue-qa
This dataset is a remastered version prepared using Adaption's Adaptive Data platform.
Marketing Spend & Revenue QA
A synthetic, table-grounded question-answering dataset for marketing analytics. Every
question is answered from an underlying marketing report (channel/segment spend, revenue,
conversions, and funnel metrics), and every answer is exact and verified by construction
not produced by a model reading a chart, but derived directly from the data the report was
built… See the full description on the dataset page: https://huggingface.co/datasets/vinod-anbalagan/adaption-marketing-spend-revenue-qa.single-turn-eval-int_qwen3-4b_distill_teacher_reverse_kl_lr1e-7-n32
Single-turn eval — violetxi/int_qwen3-4b_distill_teacher_reverse_kl_lr1e-7
Generated by teaching/inference/single_turn_eval_vllm.py. One row per problem; samples is the list of model responses, scores is per-sample correctness, and mean/best/worst are the aggregates used by mean@N / best@N / worst@N.
Eval results (n_samples_per_example = 32)
Overall
metric
value
n_examples
566
mean@32
0.3146
best@32
0.5883
worst@32
0.0919
pass_rate… See the full description on the dataset page: https://huggingface.co/datasets/PS-098/single-turn-eval-int_qwen3-4b_distill_teacher_reverse_kl_lr1e-7-n32.single-turn-eval-int_qwen3-4b_distill_teacher_reverse_kl_lr1e-7-n32
Single-turn eval — violetxi/int_qwen3-4b_distill_teacher_reverse_kl_lr1e-7
Generated by teaching/inference/single_turn_eval_vllm.py. One row per problem; samples is the list of model responses, scores is per-sample correctness, and mean/best/worst are the aggregates used by mean@N / best@N / worst@N.
Eval results (n_samples_per_example = 32)
Overall
metric
value
n_examples
566
mean@32
0.3114
best@32
0.5795
worst@32
0.0777
pass_rate… See the full description on the dataset page: https://huggingface.co/datasets/violetxi/single-turn-eval-int_qwen3-4b_distill_teacher_reverse_kl_lr1e-7-n32.price-reversal
Price Reversal Benchmark
This dataset accompanies the anonymous NeurIPS 2026 submission:
“The Price Reversal Phenomenon: When Cheaper Reasoning Models End Up Costing More”
The benchmark studies the mismatch between listed API pricing and actual inference cost in reasoning language models.
Dataset Description
The dataset contains model-level and example-level cost auditing results across single-turn reasoning benchmarks and multi-turn agentic benchmarks.
Each result… See the full description on the dataset page: https://huggingface.co/datasets/price-reversal/price-reversal.reverse-mortgage-scenarios
Reverse Mortgage (HECM) Scenarios 2026
Pre-calculated reverse mortgage scenarios for 102 combinations of age (62-85), home value ($200K-$750K), and existing mortgage balance. Includes principal limits, payment options, and heir considerations.
2026 HECM Limit
Maximum claim amount: $1,149,825
Key Benefits
No monthly mortgage payments required
You retain full ownership
Non-recourse protection
Tax-free proceeds
Expert
Wendy Thompson, CDLP
reverse-mortgage-scenarios
Reverse Mortgage Scenarios
260 HECM reverse mortgage scenarios by age, value, existing mortgage.
Details
Records: 260
Format: JSONL
License: CC-BY-4.0
Last Updated: March 2026
Verified By: Wendy Thompson, CPA, CDLP, NMLS #504814
Publisher: Wendy Thompson Lending Team
Thompson Alpha Logic
Integrates HUD Principal Limit Factor tables with Thompson 'Sequence of Returns' modeling to show how HECM line-of-credit access hedges 401(k) depletion during market… See the full description on the dataset page: https://huggingface.co/datasets/Wendy-Thompson-Lending-Team/reverse-mortgage-scenarios.FoVer-Data-Reveal
Reveal: A Benchmark for Verifiers of Reasoning Chains
Paper: A Chain-of-Thought Is as Strong as Its Weakest Link: A Benchmark for Verifiers of Reasoning Chains
Link: https://arxiv.org/abs/2402.00559
Website: https://reveal-dataset.github.io/
Abstract:
Prompting language models to provide step-by-step answers (e.g., "Chain-of-Thought") is the prominent approach for complex reasoning tasks, where more accurate reasoning chains typically improve downstream task… See the full description on the dataset page: https://huggingface.co/datasets/peiyu-cn/FoVer-Data-Reveal.