| rank,checkpoint,eval_accuracy,toxicity_rate,latency_ms,core_reasoning_avg,language_understanding_avg,generation_tasks_avg,specialized_capabilities_avg,robustness_floor,risk_adjusted_score,weakest_group,model_sha256_12 |
| 1,checkpoints/step_800,0.752,0.018,185.000,0.719,0.754,0.686,0.774,0.686,0.672,Generation Tasks,01934db10231 |
| 2,checkpoints/step_700,0.739,0.016,175.000,0.716,0.751,0.682,0.771,0.682,0.669,Generation Tasks,beeb0044e3ea |
| 3,checkpoints/step_600,0.716,0.019,205.000,0.712,0.750,0.678,0.766,0.678,0.663,Generation Tasks,177138f2d2c3 |
| 4,checkpoints/step_900,0.771,0.033,260.000,0.726,0.756,0.677,0.777,0.677,0.656,Generation Tasks,d837916e96de |
| 5,checkpoints/step_500,0.691,0.021,210.000,0.704,0.736,0.670,0.750,0.670,0.654,Generation Tasks,49e27938931c |
| 6,checkpoints/step_400,0.668,0.022,230.000,0.696,0.722,0.657,0.733,0.657,0.640,Generation Tasks,71780498800f |
| 7,checkpoints/step_300,0.641,0.024,240.000,0.684,0.710,0.644,0.721,0.644,0.626,Generation Tasks,f549bc42f96d |
| 8,checkpoints/step_200,0.612,0.028,260.000,0.672,0.695,0.630,0.705,0.630,0.610,Generation Tasks,8552743bd231 |
|
|
