{"benchmark": {"slug": "harnessrl-2699", "name": "HarnessRL-2699", "description": "HarnessRL-2699 is the OpenSWE-derived training index used for Harbor RL experiments (2,699 software-engineering tasks). Category: training \u2014 not a public leaderboard benchmark.\n\nDataset: Hugging Face `Lego-X/HarnessRL-2699`. Used in HarnessRL train configs for rollouts and verifier rewards.", "category": "training", "split": null, "language": "python", "task_count": 2699, "homepage_url": "https://huggingface.co/datasets/Lego-X/HarnessRL-2699", "paper_url": null, "hf_dataset_id": "Lego-X/HarnessRL-2699", "official_leaderboard_url": null, "is_public_leaderboard": false, "harness_count": 0, "model_count": 0, "score_count": 0, "best_success_rate": null}, "protocol": null, "rankings": {"benchmark": {"slug": "harnessrl-2699", "name": "HarnessRL-2699", "description": "HarnessRL-2699 is the OpenSWE-derived training index used for Harbor RL experiments (2,699 software-engineering tasks). Category: training \u2014 not a public leaderboard benchmark.\n\nDataset: Hugging Face `Lego-X/HarnessRL-2699`. Used in HarnessRL train configs for rollouts and verifier rewards.", "category": "training", "split": null, "language": "python", "task_count": 2699, "homepage_url": "https://huggingface.co/datasets/Lego-X/HarnessRL-2699", "paper_url": null, "hf_dataset_id": "Lego-X/HarnessRL-2699", "official_leaderboard_url": null, "is_public_leaderboard": false, "harness_count": null, "model_count": null, "score_count": null, "best_success_rate": null}, "sort": "success_rate", "rows": []}}