datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
Snowball-67B-A2B-Mixed-RLVR-Experiment-Artifacts
Snowball 67B-A2B RL artifact release
2026 mixed-domain RLVR campaign
This release also contains the complete releasable record of the September 2026 Snowball mixed-domain RLVR campaign.
It covers the September 11 synchronous and bounded-staleness asynchronous RLVR1→RLVR2 lineages and the 5.7T
Agentic-start RLVR1 lineage. All training arms are terminal. The final campaign figure,
trace audit, canonical configs, timing reports, retained
traces, and operational… See the full description on the dataset page: https://huggingface.co/datasets/open-athena/Snowball-67B-A2B-Mixed-RLVR-Experiment-Artifacts.details_yam-peleg__Experiment9-7B
Dataset Card for Evaluation run of yam-peleg/Experiment9-7B
Dataset automatically created during the evaluation run of model yam-peleg/Experiment9-7B on the Open LLM Leaderboard.
The dataset is composed of 63 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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_yam-peleg__Experiment9-7B.details_yam-peleg__Experiment1-7B
Dataset Card for Evaluation run of yam-peleg/Experiment1-7B
Dataset automatically created during the evaluation run of model yam-peleg/Experiment1-7B on the Open LLM Leaderboard.
The dataset is composed of 63 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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_yam-peleg__Experiment1-7B.details_yam-peleg__Experiment30-7B
Dataset Card for Evaluation run of yam-peleg/Experiment30-7B
Dataset automatically created during the evaluation run of model yam-peleg/Experiment30-7B on the Open LLM Leaderboard.
The dataset is composed of 63 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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_yam-peleg__Experiment30-7B.details_yam-peleg__Experiment8-7B
Dataset Card for Evaluation run of yam-peleg/Experiment8-7B
Dataset automatically created during the evaluation run of model yam-peleg/Experiment8-7B on the Open LLM Leaderboard.
The dataset is composed of 63 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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_yam-peleg__Experiment8-7B.details_adamo1139__Yi-34B-200K-rawrr1-LORA-DPO-experimental-r3
Dataset Card for Evaluation run of adamo1139/Yi-34B-200K-rawrr1-LORA-DPO-experimental-r3
Dataset automatically created during the evaluation run of model adamo1139/Yi-34B-200K-rawrr1-LORA-DPO-experimental-r3 on the Open LLM Leaderboard.
The dataset is composed of 63 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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_adamo1139__Yi-34B-200K-rawrr1-LORA-DPO-experimental-r3.details_yam-peleg__Experiment4-7B
Dataset Card for Evaluation run of yam-peleg/Experiment4-7B
Dataset automatically created during the evaluation run of model yam-peleg/Experiment4-7B on the Open LLM Leaderboard.
The dataset is composed of 63 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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_yam-peleg__Experiment4-7B.details_yam-peleg__Experiment7-7B
Dataset Card for Evaluation run of yam-peleg/Experiment7-7B
Dataset automatically created during the evaluation run of model yam-peleg/Experiment7-7B on the Open LLM Leaderboard.
The dataset is composed of 63 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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_yam-peleg__Experiment7-7B.details_cgato__TheSpice-7b-FT-ExperimentalOrca
Dataset Card for Evaluation run of cgato/TheSpice-7b-FT-ExperimentalOrca
Dataset automatically created during the evaluation run of model cgato/TheSpice-7b-FT-ExperimentalOrca on the Open LLM Leaderboard.
The dataset is composed of 63 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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_cgato__TheSpice-7b-FT-ExperimentalOrca.details_yam-peleg__Experiment26-7B
Dataset Card for Evaluation run of yam-peleg/Experiment26-7B
Dataset automatically created during the evaluation run of model yam-peleg/Experiment26-7B on the Open LLM Leaderboard.
The dataset is composed of 63 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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_yam-peleg__Experiment26-7B.details_G-reen__EXPERIMENT-ORPO-m7b2-1-merged
Dataset Card for Evaluation run of G-reen/EXPERIMENT-ORPO-m7b2-1-merged
Dataset automatically created during the evaluation run of model G-reen/EXPERIMENT-ORPO-m7b2-1-merged on the Open LLM Leaderboard.
The dataset is composed of 63 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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_G-reen__EXPERIMENT-ORPO-m7b2-1-merged.details_NotAiLOL__Apollo-7b-orpo-Experimentaldetails_G-reen__EXPERIMENT-DPO-m7b2-1-merged
Dataset Card for Evaluation run of G-reen/EXPERIMENT-DPO-m7b2-1-merged
Dataset automatically created during the evaluation run of model G-reen/EXPERIMENT-DPO-m7b2-1-merged on the Open LLM Leaderboard.
The dataset is composed of 63 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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_G-reen__EXPERIMENT-DPO-m7b2-1-merged.details_automerger__Experiment27Pastiche-7B
Dataset Card for Evaluation run of automerger/Experiment27Pastiche-7B
Dataset automatically created during the evaluation run of model automerger/Experiment27Pastiche-7B on the Open LLM Leaderboard.
The dataset is composed of 63 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/open-llm-leaderboard-old/details_automerger__Experiment27Pastiche-7B.details_yam-peleg__Experiment20-7B
Dataset Card for Evaluation run of yam-peleg/Experiment20-7B
Dataset automatically created during the evaluation run of model yam-peleg/Experiment20-7B on the Open LLM Leaderboard.
The dataset is composed of 63 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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_yam-peleg__Experiment20-7B.details_ahxt__llama2_xs_460M_experimental
Dataset Card for Evaluation run of ahxt/llama2_xs_460M_experimental
Dataset Summary
Dataset automatically created during the evaluation run of model ahxt/llama2_xs_460M_experimental on the Open LLM Leaderboard.
The dataset is composed of 64 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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_ahxt__llama2_xs_460M_experimental.details_ChaoticNeutrals__Prima-LelantaclesV7-experimental-7b
Dataset Card for Evaluation run of ChaoticNeutrals/Prima-LelantaclesV7-experimental-7b
Dataset automatically created during the evaluation run of model ChaoticNeutrals/Prima-LelantaclesV7-experimental-7b on the Open LLM Leaderboard.
The dataset is composed of 63 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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_ChaoticNeutrals__Prima-LelantaclesV7-experimental-7b.details_cognitivecomputations__dolphin-2.8-experiment26-7b
Dataset Card for Evaluation run of cognitivecomputations/dolphin-2.8-experiment26-7b
Dataset automatically created during the evaluation run of model cognitivecomputations/dolphin-2.8-experiment26-7b on the Open LLM Leaderboard.
The dataset is composed of 63 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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_cognitivecomputations__dolphin-2.8-experiment26-7b.details_yam-peleg__Experiment15-7B
Dataset Card for Evaluation run of yam-peleg/Experiment15-7B
Dataset automatically created during the evaluation run of model yam-peleg/Experiment15-7B on the Open LLM Leaderboard.
The dataset is composed of 63 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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_yam-peleg__Experiment15-7B.details_NLUHOPOE__experiment2-cause
Dataset Card for Evaluation run of NLUHOPOE/experiment2-cause
Dataset automatically created during the evaluation run of model NLUHOPOE/experiment2-cause on the Open LLM Leaderboard.
The dataset is composed of 63 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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_NLUHOPOE__experiment2-cause.details_fionazhang__mistral-experiment-6-merge
Dataset Card for Evaluation run of fionazhang/mistral-experiment-6-merge
Dataset automatically created during the evaluation run of model fionazhang/mistral-experiment-6-merge on the Open LLM Leaderboard.
The dataset is composed of 63 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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_fionazhang__mistral-experiment-6-merge.details_liminerity__Multiverse-Experiment-slerp-7b
Dataset Card for Evaluation run of liminerity/Multiverse-Experiment-slerp-7b
Dataset automatically created during the evaluation run of model liminerity/Multiverse-Experiment-slerp-7b on the Open LLM Leaderboard.
The dataset is composed of 63 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"… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_liminerity__Multiverse-Experiment-slerp-7b.details_G-reen__EXPERIMENT-SFT-m7b2-3-merged
Dataset Card for Evaluation run of G-reen/EXPERIMENT-SFT-m7b2-3-merged
Dataset automatically created during the evaluation run of model G-reen/EXPERIMENT-SFT-m7b2-3-merged on the Open LLM Leaderboard.
The dataset is composed of 63 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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_G-reen__EXPERIMENT-SFT-m7b2-3-merged.details_cognitivecomputations__dolphin-2.8-experiment26-7b-preview
Dataset Card for Evaluation run of cognitivecomputations/dolphin-2.8-experiment26-7b-preview
Dataset automatically created during the evaluation run of model cognitivecomputations/dolphin-2.8-experiment26-7b-preview on the Open LLM Leaderboard.
The dataset is composed of 63 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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_cognitivecomputations__dolphin-2.8-experiment26-7b-preview.details_juhwanlee__experiment2-cause-v1
Dataset Card for Evaluation run of juhwanlee/experiment2-cause-v1
Dataset automatically created during the evaluation run of model juhwanlee/experiment2-cause-v1 on the Open LLM Leaderboard.
The dataset is composed of 63 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/open-llm-leaderboard-old/details_juhwanlee__experiment2-cause-v1.details_yam-peleg__Experiment27-7B
Dataset Card for Evaluation run of yam-peleg/Experiment27-7B
Dataset automatically created during the evaluation run of model yam-peleg/Experiment27-7B on the Open LLM Leaderboard.
The dataset is composed of 63 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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_yam-peleg__Experiment27-7B.details_yam-peleg__Experiment25-7B
Dataset Card for Evaluation run of yam-peleg/Experiment25-7B
Dataset automatically created during the evaluation run of model yam-peleg/Experiment25-7B on the Open LLM Leaderboard.
The dataset is composed of 63 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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_yam-peleg__Experiment25-7B.details_rwitz__experiment26-truthy-iter-0
Dataset Card for Evaluation run of rwitz/experiment26-truthy-iter-0
Dataset automatically created during the evaluation run of model rwitz/experiment26-truthy-iter-0 on the Open LLM Leaderboard.
The dataset is composed of 63 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/open-llm-leaderboard-old/details_rwitz__experiment26-truthy-iter-0.details_Heng666__EastAsia-4x7B-Moe-experiment
Dataset Card for Evaluation run of Heng666/EastAsia-4x7B-Moe-experiment
Dataset automatically created during the evaluation run of model Heng666/EastAsia-4x7B-Moe-experiment on the Open LLM Leaderboard.
The dataset is composed of 63 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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_Heng666__EastAsia-4x7B-Moe-experiment.details_yam-peleg__Experiment28-7B
Dataset Card for Evaluation run of yam-peleg/Experiment28-7B
Dataset automatically created during the evaluation run of model yam-peleg/Experiment28-7B on the Open LLM Leaderboard.
The dataset is composed of 63 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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_yam-peleg__Experiment28-7B.