CoolFace
30 shown

datasets

Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.

modalityAllTextImageAudioTabularVideo
taskAllText generationQuestion answeringImage classificationSpeech
sizeAllUnder 1K1K – 10K100K – 1M1M – 10M
formatAllParquetJSONCSVWebDataset
sortTrendingMost likedMost usedRecently updatedNewest
Clear all
01LeoZotos /math500tabularn<1K0 likes715 downloads1mo agoHugging Face02alessiodevoto /math500tabularn<1K0 likes657 downloads2mo agoHugging Face03ChrisMcCormick /math500-cot-deepseek-r1-1.5b MATH-500 CoT completions (DeepSeek-R1-Distill-Qwen-1.5B) Successful chain-of-thought completions for HuggingFaceH4/MATH-500 test problems, generated with deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B via vLLM. Files File Description records.parquet Main dataset: correct completions as token IDs manifest.json Schema, tokenizer, run ids, decoding config problem_index.json unique_id → problem_idx in MATH-500 test subject_max_tokens.json Per-subject completion… See the full description on the dataset page: https://huggingface.co/datasets/ChrisMcCormick/math500-cot-deepseek-r1-1.5b.tabulartext-generation1K<n<10K0 likes72 downloads4mo agoHugging Face04lulululuyi /R-HORIZON-Math500tabular1K<n<10K0 likes69 downloads1y agoHugging Face05meituan-longcat /R-HORIZON-Math500 R-HORIZON How Far Can Your Large Reasoning Model Really Go in Breadth and Depth? 📃 Paper • 🌐 Project Page • 🤗 Dataset R-HORIZON is a novel method designed to stimulate long-horizon reasoning behaviors in Large Reasoning Models (LRMs) through query composition. We transform isolated problems into complex multi-step reasoning scenarios, revealing that even the most advanced LRMs suffer significant performance degradation when facing interdependent problems that span… See the full description on the dataset page: https://huggingface.co/datasets/meituan-longcat/R-HORIZON-Math500.tabular1K<n<10K1 likes67 downloads11mo agoHugging Face06sibasmarakp /Qwen2.5-1.5B-Instruct-uPRM-70B-T80-math500-best_of_n-completionstabular1K<n<10K0 likes61 downloads24d agoHugging Face07sibasmarakp /Llama-3.2-1B-Instruct-uPRM-32B-T80-math500-best_of_n-completionstabular1K<n<10K0 likes54 downloads24d agoHugging Face08sibasmarakp /Llama-3.2-1B-Instruct-uPRM-70B-T80-math500-best_of_n-completionstabular1K<n<10K0 likes53 downloads24d agoHugging Face09sibasmarakp /Qwen2.5-1.5B-Instruct-uPRM-32B-T80-math500-best_of_n-completionstabular1K<n<10K0 likes47 downloads24d agoHugging Face10amphora /m-math500tabularn<1K0 likes40 downloads2y agoHugging Face11osieosie /mixed_sft_math500_128_s1_tulu2_sft_s1_1.0pcttabular10K<n<100K0 likes37 downloads1y agoHugging Face12chengfu0118 /Custom-Bespoke-Stratos-7B_1753961306_eval_466d_math500_skip_attn_1 chengfu0118/Custom-Bespoke-Stratos-7B_1753961306_eval_466d_math500_skip_attn_1 Precomputed model outputs for evaluation. Evaluation Results MATH500 Accuracy: 1.80% Accuracy Questions Solved Total Questions 1.80% 9 500 tabularn<1K0 likes31 downloads1y agoHugging Face13cmpatino /math500-bon-weighted-results MATH-500 Best-of-N Weighted Selection Results Dataset Description This dataset contains the results of evaluating Best-of-N weighted selection on a subset of the MATH-500 benchmark. It was created as part of a HuggingFace internship exercise exploring how test-time compute scaling with reward models can improve LLM performance on math problems. How It Was Constructed 1. Problem Selection Started from the HuggingFaceH4/MATH-500 dataset (500 problems)… See the full description on the dataset page: https://huggingface.co/datasets/cmpatino/math500-bon-weighted-results.tabulartext-generationn<1K1 likes31 downloads5mo agoHugging Face14idoazou /math500-typostabular10K<n<100K0 likes31 downloads2mo agoHugging Face15mikasenghaas /math500-checktabularn<1K0 likes28 downloads1y agoHugging Face16albertge /MATH-500-gsm8k-format MATH-500-gsm8k-format Dataset Description This dataset contains 500 mathematical problems from the MATH-500 benchmark, converted to GSM8K format for step-by-step reasoning. Dataset Summary Source: HuggingFaceH4/MATH-500 Format: GSM8K-style step-by-step solutions with inline computation annotations Size: 500 problems Split: Test (original MATH-500 test split) Conversion Process The original MATH-500 solutions (which use LaTeX notation and… See the full description on the dataset page: https://huggingface.co/datasets/albertge/MATH-500-gsm8k-format.tabulartext-generationn<1K0 likes28 downloads11mo agoHugging Face17chengfu0118 /Unroll-Qwen2.5-7B-Instruct_1754691065_eval_6a28_math500_top-5-voting_num_prune_attn_6 chengfu0118/Unroll-Qwen2.5-7B-Instruct_1754691065_eval_6a28_math500_top-5-voting_num_prune_attn_6 Precomputed model outputs for evaluation. Evaluation Results MATH500 Accuracy: 54.80% Accuracy Questions Solved Total Questions 54.80% 274 500 tabularn<1K0 likes26 downloads1y agoHugging Face18reasoningMIA /OpenThoughts3-10k-dedup_cont3_without_math500tabular10K<n<100K0 likes24 downloads1y agoHugging Face19mikasenghaas /math500-rubric-parser-math-verifytabularn<1K0 likes24 downloads1y agoHugging Face20augustoFranke /post-training-takehome-math500-bon16 MATH-500 Best-of-16 Post-Training Take-Home Results A 50-problem study of test-time compute, based on the Hugging Face post-training take-home challenge. Nothing here trains or modifies a model: both the generator and the reward model stay frozen, and the only variable is how a final answer is chosen from 16 sampled candidates. Construction Filtered MATH-500 to levels 1-3, shuffled with seed 1, and selected 50 rows. Generated one greedy solution per problem with… See the full description on the dataset page: https://huggingface.co/datasets/augustoFranke/post-training-takehome-math500-bon16.tabulartext-generationn<1K0 likes24 downloads2mo agoHugging Face21riturajj /MATH500-Artabularn<1K0 likes23 downloads1y agoHugging Face22chengfu0118 /Custom-OpenThinker-32B_1754029333_eval_466d_math500_prn_attn_7 chengfu0118/Custom-OpenThinker-32B_1754029333_eval_466d_math500_prn_attn_7 Precomputed model outputs for evaluation. Evaluation Results MATH500 Accuracy: 0.00% Accuracy Questions Solved Total Questions 0.00% 0 500 tabularn<1K0 likes22 downloads1y agoHugging Face23chengfu0118 /Unroll-Qwen2.5-7B-Instruct_1754615103_eval_6a28_math500_skip_attn_idx_10 chengfu0118/Unroll-Qwen2.5-7B-Instruct_1754615103_eval_6a28_math500_skip_attn_idx_10 Precomputed model outputs for evaluation. Evaluation Results MATH500 Accuracy: 70.80% Accuracy Questions Solved Total Questions 70.80% 354 500 tabularn<1K0 likes22 downloads1y agoHugging Face24chengfu0118 /Unroll-Qwen2.5-7B-Instruct_1754640953_eval_6a28_math500_skip_ffn_idx_1 chengfu0118/Unroll-Qwen2.5-7B-Instruct_1754640953_eval_6a28_math500_skip_ffn_idx_1 Precomputed model outputs for evaluation. Evaluation Results MATH500 Accuracy: 70.40% Accuracy Questions Solved Total Questions 70.40% 352 500 tabularn<1K0 likes21 downloads1y agoHugging Face25chengfu0118 /Unroll-Qwen2.5-7B-Instruct_1754615061_eval_6a28_math500_skip_attn_idx_9 chengfu0118/Unroll-Qwen2.5-7B-Instruct_1754615061_eval_6a28_math500_skip_attn_idx_9 Precomputed model outputs for evaluation. Evaluation Results MATH500 Accuracy: 67.80% Accuracy Questions Solved Total Questions 67.80% 339 500 tabularn<1K0 likes19 downloads1y agoHugging Face26chengfu0118 /Unroll-Qwen2.5-7B-Instruct_1754916227_eval_6a28_math500_geometric_num_prune_ffn_4_run-002 chengfu0118/Unroll-Qwen2.5-7B-Instruct_1754916227_eval_6a28_math500_geometric_num_prune_ffn_4_run-002 Precomputed model outputs for evaluation. Evaluation Results MATH500 Accuracy: 38.20% Accuracy Questions Solved Total Questions 38.20% 191 500 tabularn<1K0 likes19 downloads1y agoHugging Face27hzy /20250317-math500-sampling-solutions-32-temptabular1K<n<10K0 likes18 downloads2y agoHugging Face28mikasenghaas /math500-rubric-math-verifytabularn<1K0 likes18 downloads1y agoHugging Face29mikasenghaas /Qwen2.5-7B-SFT-Math-Code-1M-1000-MATH500tabularn<1K0 likes18 downloads1y agoHugging Face30vibhuiitj /math500-output-audittabularn<1K0 likes18 downloads5mo agoHugging Face

Listings come live from the Hugging Face Hub API. CoolFace does not host these files.