datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
solar-flare-hmivideo-datasplitsThis dataset is intended to be used for training/testing solar flare forecasting models.
It contains various data splits (in json format) of SDO/HMI magnetogram images compiled by Boucheron, L.E., et al., 2023, Sci Data 10, 825, https://doi.org/10.1038/s41597-023-02628-8 and arranged in 16 video frames with different frame cadence (32 min, 76 min).
Splits "train_72min" ("train_36min"), "val_72min" ("val_36min"), "test_72min" ("val_36min") corresponds to the original data splits provided by… See the full description on the dataset page: https://huggingface.co/datasets/inaf-oact-ai/solar-flare-hmivideo-datasplits.upstage__SOLAR-10.7B-v1.0-details
Dataset Card for Evaluation run of upstage/SOLAR-10.7B-v1.0
Dataset automatically created during the evaluation run of model upstage/SOLAR-10.7B-v1.0
The dataset is composed of 83 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 23 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/upstage__SOLAR-10.7B-v1.0-details.solar-inverter-panel-battery-compatibility
Solar inverter–battery compatibility
Canonical, always-current version: https://referencesource.org/solar-inverter-panel-battery-compatibility/
Machine-readable: https://referencesource.org/solar-inverter-panel-battery-compatibility/data.json — this mirror is a point-in-time copy.
Last verified: 2026-08-12
Stale after: 2027-02-08 (past this date, prefer the canonical copy —
it re-verifies on a cadence this snapshot does not)
Records: 101
Which lithium batteries have been… See the full description on the dataset page: https://huggingface.co/datasets/referencesource/solar-inverter-panel-battery-compatibility.solarVQA
SolarVQA: A Benchmark for Visual Question Answering in Photovoltaic Defect Inspection
Dataset Summary
SolarVQA is a structured Visual Question Answering dataset built from
expert-annotated electroluminescence (EL) images of silicon solar cells.
It contains 130,712 QA pairs across 16,339 images spanning eight
complementary question types designed to probe defect existence, counting,
type identification, severity, localisation, co-occurrence, and spatial
distribution.… See the full description on the dataset page: https://huggingface.co/datasets/masked-token/solarVQA.upstage__solar-pro-preview-instruct-details
Dataset Card for Evaluation run of upstage/solar-pro-preview-instruct
Dataset automatically created during the evaluation run of model upstage/solar-pro-preview-instruct
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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/upstage__solar-pro-preview-instruct-details.upstage__SOLAR-10.7B-Instruct-v1.0-details
Dataset Card for Evaluation run of upstage/SOLAR-10.7B-Instruct-v1.0
Dataset automatically created during the evaluation run of model upstage/SOLAR-10.7B-Instruct-v1.0
The dataset is composed of 44 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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/upstage__SOLAR-10.7B-Instruct-v1.0-details.solar-flare-hmi-datasplitsThis dataset is intended to be used for training/testing solar flare forecasting models. It contains various data splits (in json format) of SDO/HMI magnetogram images compiled by
Boucheron, L.E., et al., 2023, Sci Data 10, 825, https://doi.org/10.1038/s41597-023-02628-8.
Splits "train", "val", "test" corresponds to the original data splits provided by Boucheron et al., while the other splits are created by downsampling the No-Flare and C flare class to obtain
more balanced splits and… See the full description on the dataset page: https://huggingface.co/datasets/inaf-oact-ai/solar-flare-hmi-datasplits.NousResearch__Yarn-Solar-10b-32k-details
Dataset Card for Evaluation run of NousResearch/Yarn-Solar-10b-32k
Dataset automatically created during the evaluation run of model NousResearch/Yarn-Solar-10b-32k
The dataset is composed of 44 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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/NousResearch__Yarn-Solar-10b-32k-details.NousResearch__Yarn-Solar-10b-64k-details
Dataset Card for Evaluation run of NousResearch/Yarn-Solar-10b-64k
Dataset automatically created during the evaluation run of model NousResearch/Yarn-Solar-10b-64k
The dataset is composed of 44 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/NousResearch__Yarn-Solar-10b-64k-details.NousResearch__Nous-Hermes-2-SOLAR-10.7B-details
Dataset Card for Evaluation run of NousResearch/Nous-Hermes-2-SOLAR-10.7B
Dataset automatically created during the evaluation run of model NousResearch/Nous-Hermes-2-SOLAR-10.7B
The dataset is composed of 44 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/open-llm-leaderboard/NousResearch__Nous-Hermes-2-SOLAR-10.7B-details.alibi-interrogation-dataset
Alibi Interrogation Dataset
1,736 synthetic multi-turn interrogation conversations for fine-tuning LLMs to play murder suspects. Built for the "Alibi" interrogation game concept.
Format
ShareGPT-style JSONL. Each line is a conversation:
{"conversations": [{"from": "system", "value": "Character card..."}, {"from": "human", "value": "Detective question"}, {"from": "gpt", "value": "Suspect response with *physical cues*"}, ...]}
Compatible with Unsloth, Axolotl, and any… See the full description on the dataset page: https://huggingface.co/datasets/solarkyle/alibi-interrogation-dataset.freewheelin__free-solar-evo-v0.13-details
Dataset Card for Evaluation run of freewheelin/free-solar-evo-v0.13
Dataset automatically created during the evaluation run of model freewheelin/free-solar-evo-v0.13
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/freewheelin__free-solar-evo-v0.13-details.solar-flare-goes-datasetThis dataset is intended to be used for training/testing solar flare forecasting models. It contains various data splits (in json format) of
GOES XRS time series (1 min-cadence) for two variables:
L2 flux/bkg ratio
Flare binary history (0=no flare, 1=flare)
Splits labelled as "_24h" correspond to a time series length of 24h, while those labelled as "_12h" to a length of 12h.
SKZF6Jq9y0K67iCmVAGOsolutions__SauerkrautLM-SOLAR-Instruct-details
Dataset Card for Evaluation run of VAGOsolutions/SauerkrautLM-SOLAR-Instruct
Dataset automatically created during the evaluation run of model VAGOsolutions/SauerkrautLM-SOLAR-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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/VAGOsolutions__SauerkrautLM-SOLAR-Instruct-details.Weyaxi__SauerkrautLM-UNA-SOLAR-Instruct-details
Dataset Card for Evaluation run of Weyaxi/SauerkrautLM-UNA-SOLAR-Instruct
Dataset automatically created during the evaluation run of model Weyaxi/SauerkrautLM-UNA-SOLAR-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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/Weyaxi__SauerkrautLM-UNA-SOLAR-Instruct-details.zKS0qsmMY9161NhYV3Eez69swvOO9tGKLmU12QfppBWgkdfOcOhzjxVcSQvqyPIehdf1O8RBVhVkXk4fM0DvEkqoErTyW1oOw4r10ck__SOLAR-10.7B-Instruct-v1.0-uncensored-details
Dataset Card for Evaluation run of w4r10ck/SOLAR-10.7B-Instruct-v1.0-uncensored
Dataset automatically created during the evaluation run of model w4r10ck/SOLAR-10.7B-Instruct-v1.0-uncensored
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/w4r10ck__SOLAR-10.7B-Instruct-v1.0-uncensored-details.bgkpwTXxtDP57KxKYxoxfSrSVUM6N19Z20flt6GNYmZx4tdFred-solar-sky-all-songs-instruct8kjfR0A42DurBxoTemJfD6GoSXw0QIigQXOKcvknpnkUvVqf