datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
3_4_fusechat_v1_openchat-3.5_mixtral-8x7b-instruct-v0.1_solar-10.7b-instruct-v1.0_representationopenchat_sharegpt_v3ShareGPT dataset for training OpenChat V3 series. See OpenChat repository for instructions.
Contents:
sharegpt_clean.json: ShareGPT dataset in original format, converted to Markdown, and with model labels.
sharegpt_gpt4.json: All instances in sharegpt_clean.json with model == "Model: GPT-4".
*.parquet: Pre-tokenized dataset for training specified version of OpenChat.
Note: The dataset is NOT currently compatible with HF dataset loader.
Licensed under MIT.
lm-eval-results-openchat-openchat-3.6-8b-20240522-private
Dataset Card for Evaluation run of openchat/openchat-3.6-8b-20240522
Dataset automatically created during the evaluation run of model openchat/openchat-3.6-8b-20240522
The dataset is composed of 62 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 4 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/nyu-dice-lab/lm-eval-results-openchat-openchat-3.6-8b-20240522-private.2_4_fusechat_v1_openchat-3.5_mixtral-8x7b-instruct-v0.1_solar-10.7b-instruct-v1.0_representation1_4_fusechat_v1_openchat-3.5_mixtral-8x7b-instruct-v0.1_solar-10.7b-instruct-v1.0_representationPretergeek__OpenChat-3.5-0106_8.99B_40Layers-Appended-details
Dataset Card for Evaluation run of Pretergeek/OpenChat-3.5-0106_8.99B_40Layers-Appended
Dataset automatically created during the evaluation run of model Pretergeek/OpenChat-3.5-0106_8.99B_40Layers-Appended
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"… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/Pretergeek__OpenChat-3.5-0106_8.99B_40Layers-Appended-details.Pretergeek__openchat-3.5-0106_Rebased_Mistral-7B-v0.2-details
Dataset Card for Evaluation run of Pretergeek/openchat-3.5-0106_Rebased_Mistral-7B-v0.2
Dataset automatically created during the evaluation run of model Pretergeek/openchat-3.5-0106_Rebased_Mistral-7B-v0.2
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"… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/Pretergeek__openchat-3.5-0106_Rebased_Mistral-7B-v0.2-details.openchat__openchat-3.5-1210-details
Dataset Card for Evaluation run of openchat/openchat-3.5-1210
Dataset automatically created during the evaluation run of model openchat/openchat-3.5-1210
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 results.
An… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/openchat__openchat-3.5-1210-details.openchat__openchat_3.5-details
Dataset Card for Evaluation run of openchat/openchat_3.5
Dataset automatically created during the evaluation run of model openchat/openchat_3.5
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 results.
An additional… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/openchat__openchat_3.5-details.openchat__openchat-3.5-0106-details
Dataset Card for Evaluation run of openchat/openchat-3.5-0106
Dataset automatically created during the evaluation run of model openchat/openchat-3.5-0106
The dataset is composed of 44 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 5 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/openchat__openchat-3.5-0106-details.openchat__openchat-3.6-8b-20240522-details
Dataset Card for Evaluation run of openchat/openchat-3.6-8b-20240522
Dataset automatically created during the evaluation run of model openchat/openchat-3.6-8b-20240522
The dataset is composed of 43 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 3 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/openchat__openchat-3.6-8b-20240522-details.Pretergeek__OpenChat-3.5-0106_32K-PoSE-details
Dataset Card for Evaluation run of Pretergeek/OpenChat-3.5-0106_32K-PoSE
Dataset automatically created during the evaluation run of model Pretergeek/OpenChat-3.5-0106_32K-PoSE
The dataset is composed of 38 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 3 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/Pretergeek__OpenChat-3.5-0106_32K-PoSE-details.openchat__openchat_v3.2_super-details
Dataset Card for Evaluation run of openchat/openchat_v3.2_super
Dataset automatically created during the evaluation run of model openchat/openchat_v3.2_super
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 results.
An… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/openchat__openchat_v3.2_super-details.beowolx__CodeNinja-1.0-OpenChat-7B-details
Dataset Card for Evaluation run of beowolx/CodeNinja-1.0-OpenChat-7B
Dataset automatically created during the evaluation run of model beowolx/CodeNinja-1.0-OpenChat-7B
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/beowolx__CodeNinja-1.0-OpenChat-7B-details.Pretergeek__OpenChat-3.5-0106_9.86B_44Layers-Appended-details
Dataset Card for Evaluation run of Pretergeek/OpenChat-3.5-0106_9.86B_44Layers-Appended
Dataset automatically created during the evaluation run of model Pretergeek/OpenChat-3.5-0106_9.86B_44Layers-Appended
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"… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/Pretergeek__OpenChat-3.5-0106_9.86B_44Layers-Appended-details.openchat__openchat_v3.2-details
Dataset Card for Evaluation run of openchat/openchat_v3.2
Dataset automatically created during the evaluation run of model openchat/openchat_v3.2
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 results.
An additional… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/openchat__openchat_v3.2-details.Pretergeek__OpenChat-3.5-0106_8.11B_36Layers-Appended-details
Dataset Card for Evaluation run of Pretergeek/OpenChat-3.5-0106_8.11B_36Layers-Appended
Dataset automatically created during the evaluation run of model Pretergeek/OpenChat-3.5-0106_8.11B_36Layers-Appended
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"… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/Pretergeek__OpenChat-3.5-0106_8.11B_36Layers-Appended-details.Pretergeek__OpenChat-3.5-0106_10.7B_48Layers-Appended-details
Dataset Card for Evaluation run of Pretergeek/OpenChat-3.5-0106_10.7B_48Layers-Appended
Dataset automatically created during the evaluation run of model Pretergeek/OpenChat-3.5-0106_10.7B_48Layers-Appended
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"… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/Pretergeek__OpenChat-3.5-0106_10.7B_48Layers-Appended-details.Pretergeek__OpenChat-3.5-0106_8.99B_40Layers-Interleaved-details
Dataset Card for Evaluation run of Pretergeek/OpenChat-3.5-0106_8.99B_40Layers-Interleaved
Dataset automatically created during the evaluation run of model Pretergeek/OpenChat-3.5-0106_8.99B_40Layers-Interleaved
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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/Pretergeek__OpenChat-3.5-0106_8.99B_40Layers-Interleaved-details.Pretergeek__OpenChat-3.5-0106_10.7B_48Layers-Interleaved-details
Dataset Card for Evaluation run of Pretergeek/OpenChat-3.5-0106_10.7B_48Layers-Interleaved
Dataset automatically created during the evaluation run of model Pretergeek/OpenChat-3.5-0106_10.7B_48Layers-Interleaved
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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/Pretergeek__OpenChat-3.5-0106_10.7B_48Layers-Interleaved-details.Pretergeek__OpenChat-3.5-0106_8.11B_36Layers-Interleaved-details
Dataset Card for Evaluation run of Pretergeek/OpenChat-3.5-0106_8.11B_36Layers-Interleaved
Dataset automatically created during the evaluation run of model Pretergeek/OpenChat-3.5-0106_8.11B_36Layers-Interleaved
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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/Pretergeek__OpenChat-3.5-0106_8.11B_36Layers-Interleaved-details.OpenChatData
OpenChatData
OpenChatData is an anonymized dataset derived from database dumps from a discontinued AI chatbot service that routed model requests through OpenRouter.
The dataset contains 20,949 chat-log records collected between February 4, 2026 and April 5, 2026, covering usage across 27 model identifiers.
Important: OpenChatData does not contain the text of user prompts or model responses. The released data consists of metadata and aggregate measurements such as token, word… See the full description on the dataset page: https://huggingface.co/datasets/gptforfree/OpenChatData.nectar_openchat_preprocessopenchat__openchat-3.5-1210nectar_openchat_preprocess2openchat3.5details_openchat__openchat-3.6-8b-20240522
Dataset Card for Evaluation run of openchat/openchat-3.6-8b-20240522
Dataset automatically created during the evaluation run of model openchat/openchat-3.6-8b-20240522.
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_openchat__openchat-3.6-8b-20240522.details_cognitivecomputations__openchat-3.5-0106-laser
Dataset Card for Evaluation run of cognitivecomputations/openchat-3.5-0106-laser
Dataset automatically created during the evaluation run of model cognitivecomputations/openchat-3.5-0106-laser.
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… See the full description on the dataset page: https://huggingface.co/datasets/OALL/details_cognitivecomputations__openchat-3.5-0106-laser.details_beowolx__CodeNinja-1.0-OpenChat-7B
Dataset Card for Evaluation run of beowolx/CodeNinja-1.0-OpenChat-7B
Dataset automatically created during the evaluation run of model beowolx/CodeNinja-1.0-OpenChat-7B.
The dataset is composed of 136 configuration, each one coresponding 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/OALL/details_beowolx__CodeNinja-1.0-OpenChat-7B.openchat__openchat_3.5