datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
Phi4-ensemble-teacher-forcing-record-logits-dataphi-4-eval-logs-and-scoresEpistemeAI__DeepThinkers-Phi4-details
Dataset Card for Evaluation run of EpistemeAI/DeepThinkers-Phi4
Dataset automatically created during the evaluation run of model EpistemeAI/DeepThinkers-Phi4
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/EpistemeAI__DeepThinkers-Phi4-details.benhaotang__phi4-qwq-sky-t1-details
Dataset Card for Evaluation run of benhaotang/phi4-qwq-sky-t1
Dataset automatically created during the evaluation run of model benhaotang/phi4-qwq-sky-t1
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/benhaotang__phi4-qwq-sky-t1-details.Triangle104__Phi4-RP-o1-Ablit-details
Dataset Card for Evaluation run of Triangle104/Phi4-RP-o1-Ablit
Dataset automatically created during the evaluation run of model Triangle104/Phi4-RP-o1-Ablit
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/Triangle104__Phi4-RP-o1-Ablit-details.Quazim0t0__Phi4Basis-14B-sce-details
Dataset Card for Evaluation run of Quazim0t0/Phi4Basis-14B-sce
Dataset automatically created during the evaluation run of model Quazim0t0/Phi4Basis-14B-sce
The dataset is composed of 38 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 latest results.
An… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/Quazim0t0__Phi4Basis-14B-sce-details.phi4-function-calling-dataset-v3msmarco_hard_negatives_phi4-14bThe data was used in the paper Don't Retrieve, Generate: Prompting LLMs for Synthetic Training Data in Dense Retrieval.
Hard-Negatives generated by Phi4-14B using 10,000 passages from the MS-Marco dataset
If this dataset was useful consider citing us :)
@misc{sinha2025dontretrievegenerateprompting,
title={Don't Retrieve, Generate: Prompting LLMs for Synthetic Training Data in Dense Retrieval},
author={Aarush Sinha},
year={2025},
eprint={2504.21015}… See the full description on the dataset page: https://huggingface.co/datasets/chungimungi/msmarco_hard_negatives_phi4-14b.phi4-conversationsRaw responses generated by Phi4 , questions from alamios/Mistral-Small-24B-Instruct-2501-Conversations
Made it to use on the QwenPhi 0.5B Draft model, but the finetune did not yield much improvement, still I have generated the dataset so here is the raw data hopefully it is useful for someone.
finetome-phi4-format
FineTome-Phi4-Format
This dataset contains 15000 samples from the mlabonne/FineTome-100k dataset,
formatted specifically for Phi-4 models with the system prompt set to "detailed thinking off".
Format
Each example follows the Phi chat format:
<|im_start|>system<|im_sep|>detailed thinking off<|im_end|>
<|im_start|>user<|im_sep|>[User content]<|im_end|>
<|im_start|>assistant<|im_sep|>[Assistant response]<|im_end|>
Source
The original data comes from… See the full description on the dataset page: https://huggingface.co/datasets/ykarout/finetome-phi4-format.nvidia-nemotron-sampled-phi4-format
Nemotron Phi-4 Format Dataset
This dataset contains examples extracted from NVIDIA/Llama-Nemotron-Post-Training-Dataset, filtered and formatted for Phi-4 fine-tuning.
Dataset Details
Source: NVIDIA/Llama-Nemotron-Post-Training-Dataset
Splits: code, math, science
Filter: system_prompt="detailed thinking on", output length between 8000-10000 characters
Format: Phi-4 chat template
Size: 20000 examples
Format
Examples follow the Phi-4 chat template:… See the full description on the dataset page: https://huggingface.co/datasets/ykarout/nvidia-nemotron-sampled-phi4-format.medqa-phi4-failure-analysisThis dataset contains a comprehensive log of reasoning and answers generated by microsoft/Phi-4-mini-instruct, evaluated on medalpaca/medical_meadow_medqa (USMLE) dataset.
This dataset represents instances where model got the answer right as well as wrong. All examples includes reasoning. The inference was performed locally on Macbook (M-series) using the MLX-LM framework (The model parameters were: temp: 0.3, max_tokens: 300).
nemotron-phi4-format-datafamily_law_phi4family law algerian in phi 4 mini format
Phi4-Mini-Prose2Tags-4B-Raw-Training-DataRaw data used to train USS-Inferprise/Phi4-Mini-Prose2Tags-4B (https://huggingface.co/USS-Inferprise/Phi4-Mini-Prose2Tags-4B)
mkurman__phi4-MedIT-10B-o1-details
Dataset Card for Evaluation run of mkurman/phi4-MedIT-10B-o1
Dataset automatically created during the evaluation run of model mkurman/phi4-MedIT-10B-o1
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/mkurman__phi4-MedIT-10B-o1-details.prithivMLmods__Phi4-Super-details
Dataset Card for Evaluation run of prithivMLmods/Phi4-Super
Dataset automatically created during the evaluation run of model prithivMLmods/Phi4-Super
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/prithivMLmods__Phi4-Super-details.Undi95__Phi4-abliterated-details
Dataset Card for Evaluation run of Undi95/Phi4-abliterated
Dataset automatically created during the evaluation run of model Undi95/Phi4-abliterated
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 additional… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/Undi95__Phi4-abliterated-details.Triangle104__Phi4-RP-o1-details
Dataset Card for Evaluation run of Triangle104/Phi4-RP-o1
Dataset automatically created during the evaluation run of model Triangle104/Phi4-RP-o1
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 additional… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/Triangle104__Phi4-RP-o1-details.Quazim0t0__Phi4.Turn.R1Distill_v1.5.1-Tensors-details
Dataset Card for Evaluation run of Quazim0t0/Phi4.Turn.R1Distill_v1.5.1-Tensors
Dataset automatically created during the evaluation run of model Quazim0t0/Phi4.Turn.R1Distill_v1.5.1-Tensors
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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/Quazim0t0__Phi4.Turn.R1Distill_v1.5.1-Tensors-details.hotmailuser__Phi4-Slerp4-14B-details
Dataset Card for Evaluation run of hotmailuser/Phi4-Slerp4-14B
Dataset automatically created during the evaluation run of model hotmailuser/Phi4-Slerp4-14B
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/hotmailuser__Phi4-Slerp4-14B-details.InLegalLaySum-Phi4-Train-16KQuazim0t0__Phi4.Turn.R1Distill.16bit-details
Dataset Card for Evaluation run of Quazim0t0/Phi4.Turn.R1Distill.16bit
Dataset automatically created during the evaluation run of model Quazim0t0/Phi4.Turn.R1Distill.16bit
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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/Quazim0t0__Phi4.Turn.R1Distill.16bit-details.Quazim0t0__Math_Phi4_Reason-details
Dataset Card for Evaluation run of Quazim0t0/Math_Phi4_Reason
Dataset automatically created during the evaluation run of model Quazim0t0/Math_Phi4_Reason
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/Quazim0t0__Math_Phi4_Reason-details.traffic_law_phi4phi4testingdataset