{ "best_global_step": null, "best_metric": null, "best_model_checkpoint": null, "epoch": 0.998452810727179, "eval_steps": 500, "global_step": 242, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 1586.0, "completions/mean_length": 689.5245971679688, "completions/mean_terminated_length": 681.90380859375, "completions/min_length": 158.0, "completions/min_terminated_length": 158.0, "epoch": 0.0041258380608561115, "grad_norm": 0.12888863682746887, "kl": 0.0, "learning_rate": 1e-06, "loss": -0.0006, "num_tokens": 575726.0, "reward": 0.07209822535514832, "reward_std": 0.09710286557674408, "rewards/code_format_reward/mean": 0.01116071455180645, "rewards/code_format_reward/std": 0.10517053306102753, "rewards/curriculum_aware_reward_fn/mean": 0.06093749776482582, "rewards/curriculum_aware_reward_fn/std": 0.1447160691022873, "step": 1 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1484.0, "completions/max_terminated_length": 1484.0, "completions/mean_length": 662.607177734375, "completions/mean_terminated_length": 662.607177734375, "completions/min_length": 58.0, "completions/min_terminated_length": 58.0, "epoch": 0.008251676121712223, "grad_norm": 0.1419609785079956, "kl": 8.70823860168457e-05, "learning_rate": 1e-06, "loss": 0.0074, "num_tokens": 1121914.0, "reward": 0.08794643729925156, "reward_std": 0.10372266918420792, "rewards/code_format_reward/mean": 0.0066964286379516125, "rewards/code_format_reward/std": 0.08164843916893005, "rewards/curriculum_aware_reward_fn/mean": 0.08125000447034836, "rewards/curriculum_aware_reward_fn/std": 0.14793506264686584, "step": 2 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.004464285714285698, "completions/max_length": 4096.0, "completions/max_terminated_length": 1680.0, "completions/mean_length": 675.8035888671875, "completions/mean_terminated_length": 660.4664306640625, "completions/min_length": 241.0, "completions/min_terminated_length": 241.0, "epoch": 0.012377514182568335, "grad_norm": 0.13957782089710236, "kl": 8.046627044677734e-05, "learning_rate": 1e-06, "loss": 0.0019, "num_tokens": 1677073.0, "reward": 0.0831473246216774, "reward_std": 0.1125701516866684, "rewards/code_format_reward/mean": 0.008928571827709675, "rewards/code_format_reward/std": 0.09417349100112915, "rewards/curriculum_aware_reward_fn/mean": 0.07421875, "rewards/curriculum_aware_reward_fn/std": 0.14322684705257416, "step": 3 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 2022.0, "completions/mean_length": 716.7701416015625, "completions/mean_terminated_length": 709.2102661132812, "completions/min_length": 240.0, "completions/min_terminated_length": 240.0, "epoch": 0.016503352243424446, "grad_norm": 0.1448298543691635, "kl": 8.744001388549805e-05, "learning_rate": 1e-06, "loss": -0.0014, "num_tokens": 2278207.0, "reward": 0.0915178582072258, "reward_std": 0.12309108674526215, "rewards/code_format_reward/mean": 0.013392857275903225, "rewards/code_format_reward/std": 0.11507843434810638, "rewards/curriculum_aware_reward_fn/mean": 0.078125, "rewards/curriculum_aware_reward_fn/std": 0.14590321481227875, "step": 4 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 1974.0, "completions/mean_length": 732.2969360351562, "completions/mean_terminated_length": 724.7717895507812, "completions/min_length": 153.0, "completions/min_terminated_length": 153.0, "epoch": 0.020629190304280558, "grad_norm": 0.12635301053524017, "kl": 0.00010377168655395508, "learning_rate": 1e-06, "loss": 0.0166, "num_tokens": 2885219.0, "reward": 0.08125000447034836, "reward_std": 0.1115899309515953, "rewards/code_format_reward/mean": 0.015625, "rewards/code_format_reward/std": 0.12415824085474014, "rewards/curriculum_aware_reward_fn/mean": 0.06562500447034836, "rewards/curriculum_aware_reward_fn/std": 0.1407126635313034, "step": 5 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1418.0, "completions/max_terminated_length": 1418.0, "completions/mean_length": 659.0201416015625, "completions/mean_terminated_length": 659.0201416015625, "completions/min_length": 210.0, "completions/min_terminated_length": 210.0, "epoch": 0.02475502836513667, "grad_norm": 0.13044793903827667, "kl": 0.0001366138458251953, "learning_rate": 1e-06, "loss": 0.0077, "num_tokens": 3455763.0, "reward": 0.09564732015132904, "reward_std": 0.10143062472343445, "rewards/code_format_reward/mean": 0.008928571827709675, "rewards/code_format_reward/std": 0.09417349100112915, "rewards/curriculum_aware_reward_fn/mean": 0.08671874552965164, "rewards/curriculum_aware_reward_fn/std": 0.15126961469650269, "step": 6 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.004464285714285698, "completions/max_length": 4096.0, "completions/max_terminated_length": 2817.0, "completions/mean_length": 696.794677734375, "completions/mean_terminated_length": 681.5515747070312, "completions/min_length": 13.0, "completions/min_terminated_length": 13.0, "epoch": 0.02888086642599278, "grad_norm": 0.14978308975696564, "kl": 0.00014662742614746094, "learning_rate": 1e-06, "loss": 0.0182, "num_tokens": 4048669.0, "reward": 0.1104910746216774, "reward_std": 0.14611311256885529, "rewards/code_format_reward/mean": 0.0245535708963871, "rewards/code_format_reward/std": 0.1549331247806549, "rewards/curriculum_aware_reward_fn/mean": 0.0859375, "rewards/curriculum_aware_reward_fn/std": 0.15080992877483368, "step": 7 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1920.0, "completions/max_terminated_length": 1920.0, "completions/mean_length": 649.0826416015625, "completions/mean_terminated_length": 649.0826416015625, "completions/min_length": 235.0, "completions/min_terminated_length": 235.0, "epoch": 0.03300670448684889, "grad_norm": 0.15468838810920715, "kl": 0.0001780986785888672, "learning_rate": 1e-06, "loss": 0.0164, "num_tokens": 4602791.0, "reward": 0.0987723246216774, "reward_std": 0.1482853889465332, "rewards/code_format_reward/mean": 0.0245535708963871, "rewards/code_format_reward/std": 0.1549331247806549, "rewards/curriculum_aware_reward_fn/mean": 0.07421875, "rewards/curriculum_aware_reward_fn/std": 0.14322684705257416, "step": 8 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.004464285714285698, "completions/max_length": 4096.0, "completions/max_terminated_length": 1697.0, "completions/mean_length": 716.044677734375, "completions/mean_terminated_length": 700.887939453125, "completions/min_length": 200.0, "completions/min_terminated_length": 200.0, "epoch": 0.037132542547705004, "grad_norm": 0.14617083966732025, "kl": 0.00019991397857666016, "learning_rate": 1e-06, "loss": 0.0075, "num_tokens": 5185078.0, "reward": 0.10535714775323868, "reward_std": 0.16500398516654968, "rewards/code_format_reward/mean": 0.0334821417927742, "rewards/code_format_reward/std": 0.1800929754972458, "rewards/curriculum_aware_reward_fn/mean": 0.07187499850988388, "rewards/curriculum_aware_reward_fn/std": 0.1415448784828186, "step": 9 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1472.0, "completions/max_terminated_length": 1472.0, "completions/mean_length": 656.4732666015625, "completions/mean_terminated_length": 656.4732666015625, "completions/min_length": 92.0, "completions/min_terminated_length": 92.0, "epoch": 0.041258380608561115, "grad_norm": 0.1518249213695526, "kl": 0.000293731689453125, "learning_rate": 1e-06, "loss": 0.0105, "num_tokens": 5749623.0, "reward": 0.0993303582072258, "reward_std": 0.14284572005271912, "rewards/code_format_reward/mean": 0.02901785634458065, "rewards/code_format_reward/std": 0.16804419457912445, "rewards/curriculum_aware_reward_fn/mean": 0.0703125, "rewards/curriculum_aware_reward_fn/std": 0.14039060473442078, "step": 10 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1258.0, "completions/max_terminated_length": 1258.0, "completions/mean_length": 631.8616333007812, "completions/mean_terminated_length": 631.8616333007812, "completions/min_length": 187.0, "completions/min_terminated_length": 187.0, "epoch": 0.04538421866941723, "grad_norm": 0.16050200164318085, "kl": 0.0004885196685791016, "learning_rate": 1e-06, "loss": 0.0111, "num_tokens": 6297745.0, "reward": 0.1389508992433548, "reward_std": 0.16370971500873566, "rewards/code_format_reward/mean": 0.0334821417927742, "rewards/code_format_reward/std": 0.1800929754972458, "rewards/curriculum_aware_reward_fn/mean": 0.10546875, "rewards/curriculum_aware_reward_fn/std": 0.16077345609664917, "step": 11 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1540.0, "completions/max_terminated_length": 1540.0, "completions/mean_length": 639.546875, "completions/mean_terminated_length": 639.546875, "completions/min_length": 234.0, "completions/min_terminated_length": 234.0, "epoch": 0.04951005673027334, "grad_norm": 0.15505070984363556, "kl": 0.0007357597351074219, "learning_rate": 1e-06, "loss": 0.0142, "num_tokens": 6852201.0, "reward": 0.13794644176959991, "reward_std": 0.19176238775253296, "rewards/code_format_reward/mean": 0.0691964253783226, "rewards/code_format_reward/std": 0.25407159328460693, "rewards/curriculum_aware_reward_fn/mean": 0.06875000149011612, "rewards/curriculum_aware_reward_fn/std": 0.13920918107032776, "step": 12 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1305.0, "completions/max_terminated_length": 1305.0, "completions/mean_length": 633.4888916015625, "completions/mean_terminated_length": 633.4888916015625, "completions/min_length": 243.0, "completions/min_terminated_length": 243.0, "epoch": 0.05363589479112945, "grad_norm": 0.1798263043165207, "kl": 0.0011224746704101562, "learning_rate": 1e-06, "loss": 0.0134, "num_tokens": 7404503.0, "reward": 0.175223246216774, "reward_std": 0.2504880428314209, "rewards/code_format_reward/mean": 0.0892857164144516, "rewards/code_format_reward/std": 0.2854745090007782, "rewards/curriculum_aware_reward_fn/mean": 0.0859375, "rewards/curriculum_aware_reward_fn/std": 0.15080992877483368, "step": 13 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1794.0, "completions/max_terminated_length": 1794.0, "completions/mean_length": 629.171875, "completions/mean_terminated_length": 629.171875, "completions/min_length": 237.0, "completions/min_terminated_length": 237.0, "epoch": 0.05776173285198556, "grad_norm": 0.19240802526474, "kl": 0.0013980865478515625, "learning_rate": 1e-06, "loss": 0.0124, "num_tokens": 7971475.0, "reward": 0.19944198429584503, "reward_std": 0.27755308151245117, "rewards/code_format_reward/mean": 0.1205357164144516, "rewards/code_format_reward/std": 0.3259509205818176, "rewards/curriculum_aware_reward_fn/mean": 0.07890624552965164, "rewards/curriculum_aware_reward_fn/std": 0.14642010629177094, "step": 14 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1507.0, "completions/max_terminated_length": 1507.0, "completions/mean_length": 581.107177734375, "completions/mean_terminated_length": 581.107177734375, "completions/min_length": 168.0, "completions/min_terminated_length": 168.0, "epoch": 0.06188757091284167, "grad_norm": 0.2231186330318451, "kl": 0.0020952224731445312, "learning_rate": 1e-06, "loss": -0.0005, "num_tokens": 8486387.0, "reward": 0.2685267925262451, "reward_std": 0.32140815258026123, "rewards/code_format_reward/mean": 0.1607142835855484, "rewards/code_format_reward/std": 0.3676777780056, "rewards/curriculum_aware_reward_fn/mean": 0.10781250894069672, "rewards/curriculum_aware_reward_fn/std": 0.16176913678646088, "step": 15 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 1717.0, "completions/mean_length": 621.4464721679688, "completions/mean_terminated_length": 613.6734008789062, "completions/min_length": 253.0, "completions/min_terminated_length": 253.0, "epoch": 0.06601340897369778, "grad_norm": 0.23204894363880157, "kl": 0.002674102783203125, "learning_rate": 1e-06, "loss": 0.0265, "num_tokens": 9049387.0, "reward": 0.2908482253551483, "reward_std": 0.37603747844696045, "rewards/code_format_reward/mean": 0.1986607164144516, "rewards/code_format_reward/std": 0.3994380831718445, "rewards/curriculum_aware_reward_fn/mean": 0.09218750149011612, "rewards/curriculum_aware_reward_fn/std": 0.15433812141418457, "step": 16 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1596.0, "completions/max_terminated_length": 1596.0, "completions/mean_length": 610.296875, "completions/mean_terminated_length": 610.296875, "completions/min_length": 181.0, "completions/min_terminated_length": 181.0, "epoch": 0.07013924703455389, "grad_norm": 0.2304592877626419, "kl": 0.0031871795654296875, "learning_rate": 1e-06, "loss": 0.0171, "num_tokens": 9581863.0, "reward": 0.3455357849597931, "reward_std": 0.42431119084358215, "rewards/code_format_reward/mean": 0.2611607015132904, "rewards/code_format_reward/std": 0.43975841999053955, "rewards/curriculum_aware_reward_fn/mean": 0.08437498658895493, "rewards/curriculum_aware_reward_fn/std": 0.1552628129720688, "step": 17 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 1457.0, "completions/mean_length": 572.6875, "completions/mean_terminated_length": 564.8053588867188, "completions/min_length": 199.0, "completions/min_terminated_length": 199.0, "epoch": 0.07426508509541001, "grad_norm": 0.261919230222702, "kl": 0.004901885986328125, "learning_rate": 1e-06, "loss": 0.0376, "num_tokens": 10107808.0, "reward": 0.48091521859169006, "reward_std": 0.4869263172149658, "rewards/code_format_reward/mean": 0.3816964328289032, "rewards/code_format_reward/std": 0.4863457679748535, "rewards/curriculum_aware_reward_fn/mean": 0.09921874105930328, "rewards/curriculum_aware_reward_fn/std": 0.15791727602481842, "step": 18 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1584.0, "completions/max_terminated_length": 1584.0, "completions/mean_length": 578.529052734375, "completions/mean_terminated_length": 578.529052734375, "completions/min_length": 178.0, "completions/min_terminated_length": 178.0, "epoch": 0.07839092315626611, "grad_norm": 0.2614266276359558, "kl": 0.0062408447265625, "learning_rate": 1e-06, "loss": 0.0298, "num_tokens": 10657468.0, "reward": 0.5408482551574707, "reward_std": 0.5003149509429932, "rewards/code_format_reward/mean": 0.4799107015132904, "rewards/code_format_reward/std": 0.5001547932624817, "rewards/curriculum_aware_reward_fn/mean": 0.06093749776482582, "rewards/curriculum_aware_reward_fn/std": 0.1328689157962799, "step": 19 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1248.0, "completions/max_terminated_length": 1248.0, "completions/mean_length": 578.3683471679688, "completions/mean_terminated_length": 578.3683471679688, "completions/min_length": 226.0, "completions/min_terminated_length": 226.0, "epoch": 0.08251676121712223, "grad_norm": 0.2431550920009613, "kl": 0.00843048095703125, "learning_rate": 1e-06, "loss": 0.031, "num_tokens": 11185762.0, "reward": 0.682366132736206, "reward_std": 0.4844993054866791, "rewards/code_format_reward/mean": 0.5870535969734192, "rewards/code_format_reward/std": 0.4929138123989105, "rewards/curriculum_aware_reward_fn/mean": 0.09531249105930328, "rewards/curriculum_aware_reward_fn/std": 0.1559782177209854, "step": 20 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1274.0, "completions/max_terminated_length": 1274.0, "completions/mean_length": 565.325927734375, "completions/mean_terminated_length": 565.325927734375, "completions/min_length": 206.0, "completions/min_terminated_length": 206.0, "epoch": 0.08664259927797834, "grad_norm": 0.2159644216299057, "kl": 0.011871337890625, "learning_rate": 1e-06, "loss": 0.0291, "num_tokens": 11700009.0, "reward": 0.8511161208152771, "reward_std": 0.41165074706077576, "rewards/code_format_reward/mean": 0.7745535969734192, "rewards/code_format_reward/std": 0.41834312677383423, "rewards/curriculum_aware_reward_fn/mean": 0.07656250149011612, "rewards/curriculum_aware_reward_fn/std": 0.14485128223896027, "step": 21 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1708.0, "completions/max_terminated_length": 1708.0, "completions/mean_length": 551.1451416015625, "completions/mean_terminated_length": 551.1451416015625, "completions/min_length": 226.0, "completions/min_terminated_length": 226.0, "epoch": 0.09076843733883445, "grad_norm": 0.2016810178756714, "kl": 0.0140228271484375, "learning_rate": 1e-06, "loss": 0.004, "num_tokens": 12204053.0, "reward": 0.9275670051574707, "reward_std": 0.35804861783981323, "rewards/code_format_reward/mean": 0.8236607313156128, "rewards/code_format_reward/std": 0.3815346360206604, "rewards/curriculum_aware_reward_fn/mean": 0.10390625149011612, "rewards/curriculum_aware_reward_fn/std": 0.16008710861206055, "step": 22 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1169.0, "completions/max_terminated_length": 1169.0, "completions/mean_length": 530.3348388671875, "completions/mean_terminated_length": 530.3348388671875, "completions/min_length": 200.0, "completions/min_terminated_length": 200.0, "epoch": 0.09489427539969056, "grad_norm": 0.19738376140594482, "kl": 0.01459503173828125, "learning_rate": 1e-06, "loss": 0.0256, "num_tokens": 12704192.0, "reward": 0.9656250476837158, "reward_std": 0.2683509886264801, "rewards/code_format_reward/mean": 0.890625, "rewards/code_format_reward/std": 0.3124580383300781, "rewards/curriculum_aware_reward_fn/mean": 0.07499999552965164, "rewards/curriculum_aware_reward_fn/std": 0.14377461373806, "step": 23 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1563.0, "completions/max_terminated_length": 1563.0, "completions/mean_length": 543.46875, "completions/mean_terminated_length": 543.46875, "completions/min_length": 176.0, "completions/min_terminated_length": 176.0, "epoch": 0.09902011346054668, "grad_norm": 0.18200120329856873, "kl": 0.0171661376953125, "learning_rate": 1e-06, "loss": -0.0023, "num_tokens": 13219931.0, "reward": 1.0159599781036377, "reward_std": 0.23269660770893097, "rewards/code_format_reward/mean": 0.9308035969734192, "rewards/code_format_reward/std": 0.25407159328460693, "rewards/curriculum_aware_reward_fn/mean": 0.08515625447034836, "rewards/curriculum_aware_reward_fn/std": 0.15034477412700653, "step": 24 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1587.0, "completions/max_terminated_length": 1587.0, "completions/mean_length": 517.0982666015625, "completions/mean_terminated_length": 517.0982666015625, "completions/min_length": 198.0, "completions/min_terminated_length": 198.0, "epoch": 0.10314595152140278, "grad_norm": 0.1988343447446823, "kl": 0.0178070068359375, "learning_rate": 1e-06, "loss": 0.0037, "num_tokens": 13722008.0, "reward": 1.0328125953674316, "reward_std": 0.2289527952671051, "rewards/code_format_reward/mean": 0.9375, "rewards/code_format_reward/std": 0.24233205616474152, "rewards/curriculum_aware_reward_fn/mean": 0.09531249850988388, "rewards/curriculum_aware_reward_fn/std": 0.1559782177209854, "step": 25 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 2321.0, "completions/mean_length": 545.5960083007812, "completions/mean_terminated_length": 537.6532592773438, "completions/min_length": 131.0, "completions/min_terminated_length": 131.0, "epoch": 0.1072717895822589, "grad_norm": 0.17526358366012573, "kl": 0.0179901123046875, "learning_rate": 1e-06, "loss": 0.0148, "num_tokens": 14243754.0, "reward": 1.0184152126312256, "reward_std": 0.18650338053703308, "rewards/code_format_reward/mean": 0.9441964030265808, "rewards/code_format_reward/std": 0.22979861497879028, "rewards/curriculum_aware_reward_fn/mean": 0.07421875, "rewards/curriculum_aware_reward_fn/std": 0.14322684705257416, "step": 26 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 1612.0, "completions/mean_length": 539.9732666015625, "completions/mean_terminated_length": 532.0178833007812, "completions/min_length": 198.0, "completions/min_terminated_length": 198.0, "epoch": 0.111397627643115, "grad_norm": 0.17852729558944702, "kl": 0.019775390625, "learning_rate": 1e-06, "loss": 0.0216, "num_tokens": 14761182.0, "reward": 1.017076015472412, "reward_std": 0.20738928020000458, "rewards/code_format_reward/mean": 0.9397321343421936, "rewards/code_format_reward/std": 0.23824846744537354, "rewards/curriculum_aware_reward_fn/mean": 0.07734375447034836, "rewards/curriculum_aware_reward_fn/std": 0.14538030326366425, "step": 27 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 2059.0, "completions/mean_length": 501.2701110839844, "completions/mean_terminated_length": 493.2281799316406, "completions/min_length": 179.0, "completions/min_terminated_length": 179.0, "epoch": 0.11552346570397112, "grad_norm": 0.18558189272880554, "kl": 0.0224456787109375, "learning_rate": 1e-06, "loss": 0.034, "num_tokens": 15256959.0, "reward": 1.0671876668930054, "reward_std": 0.16758912801742554, "rewards/code_format_reward/mean": 0.96875, "rewards/code_format_reward/std": 0.17418713867664337, "rewards/curriculum_aware_reward_fn/mean": 0.09843749552965164, "rewards/curriculum_aware_reward_fn/std": 0.15753914415836334, "step": 28 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1795.0, "completions/max_terminated_length": 1795.0, "completions/mean_length": 494.08038330078125, "completions/mean_terminated_length": 494.08038330078125, "completions/min_length": 160.0, "completions/min_terminated_length": 160.0, "epoch": 0.11964930376482723, "grad_norm": 0.17248578369617462, "kl": 0.0224456787109375, "learning_rate": 1e-06, "loss": 0.0269, "num_tokens": 15755297.0, "reward": 1.0885045528411865, "reward_std": 0.10364294797182083, "rewards/code_format_reward/mean": 0.9955357313156128, "rewards/code_format_reward/std": 0.06674052774906158, "rewards/curriculum_aware_reward_fn/mean": 0.09296875447034836, "rewards/curriculum_aware_reward_fn/std": 0.15651653707027435, "step": 29 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 2559.0, "completions/mean_length": 516.46875, "completions/mean_terminated_length": 508.4608459472656, "completions/min_length": 128.0, "completions/min_terminated_length": 128.0, "epoch": 0.12377514182568335, "grad_norm": 0.1416454315185547, "kl": 0.0227203369140625, "learning_rate": 1e-06, "loss": 0.0175, "num_tokens": 16250043.0, "reward": 1.0487723350524902, "reward_std": 0.09710026532411575, "rewards/code_format_reward/mean": 0.9776785969734192, "rewards/code_format_reward/std": 0.14789186418056488, "rewards/curriculum_aware_reward_fn/mean": 0.07109375298023224, "rewards/curriculum_aware_reward_fn/std": 0.1409710794687271, "step": 30 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1677.0, "completions/max_terminated_length": 1677.0, "completions/mean_length": 496.9754638671875, "completions/mean_terminated_length": 496.9754638671875, "completions/min_length": 172.0, "completions/min_terminated_length": 172.0, "epoch": 0.12790097988653945, "grad_norm": 0.18218933045864105, "kl": 0.0243377685546875, "learning_rate": 1e-06, "loss": 0.0111, "num_tokens": 16754620.0, "reward": 1.0832589864730835, "reward_std": 0.11736033856868744, "rewards/code_format_reward/mean": 0.9910714030265808, "rewards/code_format_reward/std": 0.09417349100112915, "rewards/curriculum_aware_reward_fn/mean": 0.09218750149011612, "rewards/curriculum_aware_reward_fn/std": 0.1561036854982376, "step": 31 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2773.0, "completions/max_terminated_length": 2773.0, "completions/mean_length": 488.4151916503906, "completions/mean_terminated_length": 488.4151916503906, "completions/min_length": 185.0, "completions/min_terminated_length": 185.0, "epoch": 0.13202681794739557, "grad_norm": 0.15871264040470123, "kl": 0.0238800048828125, "learning_rate": 1e-06, "loss": 0.0051, "num_tokens": 17244008.0, "reward": 1.062611699104309, "reward_std": 0.1266116350889206, "rewards/code_format_reward/mean": 0.9821428656578064, "rewards/code_format_reward/std": 0.13258016109466553, "rewards/curriculum_aware_reward_fn/mean": 0.08046875149011612, "rewards/curriculum_aware_reward_fn/std": 0.1474359631538391, "step": 32 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1137.0, "completions/max_terminated_length": 1137.0, "completions/mean_length": 481.544677734375, "completions/mean_terminated_length": 481.544677734375, "completions/min_length": 180.0, "completions/min_terminated_length": 180.0, "epoch": 0.1361526560082517, "grad_norm": 0.18614180386066437, "kl": 0.02935791015625, "learning_rate": 1e-06, "loss": 0.011, "num_tokens": 17737381.0, "reward": 1.0718750953674316, "reward_std": 0.1312674880027771, "rewards/code_format_reward/mean": 0.984375, "rewards/code_format_reward/std": 0.12415824085474014, "rewards/curriculum_aware_reward_fn/mean": 0.08749999850988388, "rewards/curriculum_aware_reward_fn/std": 0.15172386169433594, "step": 33 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1613.0, "completions/max_terminated_length": 1613.0, "completions/mean_length": 512.8549194335938, "completions/mean_terminated_length": 512.8549194335938, "completions/min_length": 181.0, "completions/min_terminated_length": 181.0, "epoch": 0.14027849406910778, "grad_norm": 0.16082872450351715, "kl": 0.0222625732421875, "learning_rate": 1e-06, "loss": 0.0197, "num_tokens": 18228310.0, "reward": 1.0741071701049805, "reward_std": 0.12349522858858109, "rewards/code_format_reward/mean": 0.9866071343421936, "rewards/code_format_reward/std": 0.11507844179868698, "rewards/curriculum_aware_reward_fn/mean": 0.08749999850988388, "rewards/curriculum_aware_reward_fn/std": 0.15529432892799377, "step": 34 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1746.0, "completions/max_terminated_length": 1746.0, "completions/mean_length": 468.5870666503906, "completions/mean_terminated_length": 468.5870666503906, "completions/min_length": 129.0, "completions/min_terminated_length": 129.0, "epoch": 0.1444043321299639, "grad_norm": 0.1922416239976883, "kl": 0.0261993408203125, "learning_rate": 1e-06, "loss": 0.0045, "num_tokens": 18722633.0, "reward": 1.0731027126312256, "reward_std": 0.14413729310035706, "rewards/code_format_reward/mean": 0.9754464030265808, "rewards/code_format_reward/std": 0.1549331247806549, "rewards/curriculum_aware_reward_fn/mean": 0.0976562425494194, "rewards/curriculum_aware_reward_fn/std": 0.15715619921684265, "step": 35 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2882.0, "completions/max_terminated_length": 2882.0, "completions/mean_length": 519.6295166015625, "completions/mean_terminated_length": 519.6295166015625, "completions/min_length": 142.0, "completions/min_terminated_length": 142.0, "epoch": 0.14853017019082002, "grad_norm": 0.1476685255765915, "kl": 0.0240020751953125, "learning_rate": 1e-06, "loss": 0.0086, "num_tokens": 19224703.0, "reward": 1.071540355682373, "reward_std": 0.0882287248969078, "rewards/code_format_reward/mean": 0.9910714030265808, "rewards/code_format_reward/std": 0.09417349100112915, "rewards/curriculum_aware_reward_fn/mean": 0.08046875149011612, "rewards/curriculum_aware_reward_fn/std": 0.1474359631538391, "step": 36 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.004464285714285698, "completions/max_length": 4096.0, "completions/max_terminated_length": 2005.0, "completions/mean_length": 511.08929443359375, "completions/mean_terminated_length": 495.01348876953125, "completions/min_length": 183.0, "completions/min_terminated_length": 183.0, "epoch": 0.15265600825167613, "grad_norm": 0.16821761429309845, "kl": 0.02496337890625, "learning_rate": 1e-06, "loss": 0.0438, "num_tokens": 19721743.0, "reward": 1.0640625953674316, "reward_std": 0.13217082619667053, "rewards/code_format_reward/mean": 0.984375, "rewards/code_format_reward/std": 0.12415824085474014, "rewards/curriculum_aware_reward_fn/mean": 0.07968749850988388, "rewards/curriculum_aware_reward_fn/std": 0.146930992603302, "step": 37 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2589.0, "completions/max_terminated_length": 2589.0, "completions/mean_length": 505.79241943359375, "completions/mean_terminated_length": 505.79241943359375, "completions/min_length": 144.0, "completions/min_terminated_length": 144.0, "epoch": 0.15678184631253222, "grad_norm": 0.16077841818332672, "kl": 0.0255889892578125, "learning_rate": 1e-06, "loss": 0.0078, "num_tokens": 20205826.0, "reward": 1.0842634439468384, "reward_std": 0.12369215488433838, "rewards/code_format_reward/mean": 0.9866071343421936, "rewards/code_format_reward/std": 0.11507843434810638, "rewards/curriculum_aware_reward_fn/mean": 0.09765625, "rewards/curriculum_aware_reward_fn/std": 0.15715619921684265, "step": 38 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 3452.0, "completions/mean_length": 531.9263916015625, "completions/mean_terminated_length": 523.9530029296875, "completions/min_length": 150.0, "completions/min_terminated_length": 150.0, "epoch": 0.16090768437338834, "grad_norm": 0.1493963897228241, "kl": 0.0219573974609375, "learning_rate": 1e-06, "loss": 0.0287, "num_tokens": 20714770.0, "reward": 1.0684152841567993, "reward_std": 0.11183035373687744, "rewards/code_format_reward/mean": 0.9910714030265808, "rewards/code_format_reward/std": 0.09417349100112915, "rewards/curriculum_aware_reward_fn/mean": 0.07734375447034836, "rewards/curriculum_aware_reward_fn/std": 0.14538030326366425, "step": 39 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1251.0, "completions/max_terminated_length": 1251.0, "completions/mean_length": 510.1785888671875, "completions/mean_terminated_length": 510.1785888671875, "completions/min_length": 192.0, "completions/min_terminated_length": 192.0, "epoch": 0.16503352243424446, "grad_norm": 0.1577296257019043, "kl": 0.025360107421875, "learning_rate": 1e-06, "loss": 0.0097, "num_tokens": 21206971.0, "reward": 1.0684152841567993, "reward_std": 0.10139289498329163, "rewards/code_format_reward/mean": 0.9910714030265808, "rewards/code_format_reward/std": 0.09417349100112915, "rewards/curriculum_aware_reward_fn/mean": 0.07734374701976776, "rewards/curriculum_aware_reward_fn/std": 0.14538030326366425, "step": 40 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 2377.0, "completions/mean_length": 511.3594055175781, "completions/mean_terminated_length": 503.3400573730469, "completions/min_length": 166.0, "completions/min_terminated_length": 166.0, "epoch": 0.16915936049510058, "grad_norm": 0.15489205718040466, "kl": 0.027496337890625, "learning_rate": 1e-06, "loss": 0.0479, "num_tokens": 21706680.0, "reward": 1.0683035850524902, "reward_std": 0.10432274639606476, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843916893005, "rewards/curriculum_aware_reward_fn/mean": 0.07499999552965164, "rewards/curriculum_aware_reward_fn/std": 0.14377461373806, "step": 41 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1208.0, "completions/max_terminated_length": 1208.0, "completions/mean_length": 458.1094055175781, "completions/mean_terminated_length": 458.1094055175781, "completions/min_length": 163.0, "completions/min_terminated_length": 163.0, "epoch": 0.17328519855595667, "grad_norm": 0.16796942055225372, "kl": 0.0244293212890625, "learning_rate": 1e-06, "loss": 0.0076, "num_tokens": 22153287.0, "reward": 1.0933036804199219, "reward_std": 0.10752787441015244, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843171834946, "rewards/curriculum_aware_reward_fn/mean": 0.10000000149011612, "rewards/curriculum_aware_reward_fn/std": 0.15829063951969147, "step": 42 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 3074.0, "completions/max_terminated_length": 3074.0, "completions/mean_length": 500.6339416503906, "completions/mean_terminated_length": 500.6339416503906, "completions/min_length": 176.0, "completions/min_terminated_length": 176.0, "epoch": 0.1774110366168128, "grad_norm": 0.1404551863670349, "kl": 0.0234832763671875, "learning_rate": 1e-06, "loss": 0.03, "num_tokens": 22638251.0, "reward": 1.0720982551574707, "reward_std": 0.08823520690202713, "rewards/code_format_reward/mean": 0.9955357313156128, "rewards/code_format_reward/std": 0.06674052774906158, "rewards/curriculum_aware_reward_fn/mean": 0.07656250149011612, "rewards/curriculum_aware_reward_fn/std": 0.14485128223896027, "step": 43 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 1282.0, "completions/mean_length": 517.154052734375, "completions/mean_terminated_length": 509.14764404296875, "completions/min_length": 136.0, "completions/min_terminated_length": 136.0, "epoch": 0.1815368746776689, "grad_norm": 0.1547202318906784, "kl": 0.02197265625, "learning_rate": 1e-06, "loss": 0.0262, "num_tokens": 23115433.0, "reward": 1.060044765472412, "reward_std": 0.11939653754234314, "rewards/code_format_reward/mean": 0.9866071343421936, "rewards/code_format_reward/std": 0.11507843434810638, "rewards/curriculum_aware_reward_fn/mean": 0.07343750447034836, "rewards/curriculum_aware_reward_fn/std": 0.14832323789596558, "step": 44 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 1640.0, "completions/mean_length": 532.0647583007812, "completions/mean_terminated_length": 524.0917358398438, "completions/min_length": 224.0, "completions/min_terminated_length": 224.0, "epoch": 0.18566271273852503, "grad_norm": 0.1570153385400772, "kl": 0.020660400390625, "learning_rate": 1e-06, "loss": -0.0065, "num_tokens": 23643220.0, "reward": 1.0776786804199219, "reward_std": 0.10400114208459854, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843171834946, "rewards/curriculum_aware_reward_fn/mean": 0.08437500149011612, "rewards/curriculum_aware_reward_fn/std": 0.14987409114837646, "step": 45 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 3661.0, "completions/mean_length": 523.357177734375, "completions/mean_terminated_length": 515.3646850585938, "completions/min_length": 160.0, "completions/min_terminated_length": 160.0, "epoch": 0.18978855079938112, "grad_norm": 0.15556640923023224, "kl": 0.02166748046875, "learning_rate": 1e-06, "loss": 0.033, "num_tokens": 24145833.0, "reward": 1.0652903318405151, "reward_std": 0.10849537700414658, "rewards/code_format_reward/mean": 0.9910714030265808, "rewards/code_format_reward/std": 0.09417349100112915, "rewards/curriculum_aware_reward_fn/mean": 0.07421875, "rewards/curriculum_aware_reward_fn/std": 0.14322684705257416, "step": 46 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1129.0, "completions/max_terminated_length": 1129.0, "completions/mean_length": 510.8058166503906, "completions/mean_terminated_length": 510.8058166503906, "completions/min_length": 136.0, "completions/min_terminated_length": 136.0, "epoch": 0.19391438886023724, "grad_norm": 0.16910067200660706, "kl": 0.0249481201171875, "learning_rate": 1e-06, "loss": -0.0031, "num_tokens": 24644804.0, "reward": 1.0970983505249023, "reward_std": 0.11309925466775894, "rewards/code_format_reward/mean": 0.9955357313156128, "rewards/code_format_reward/std": 0.06674052774906158, "rewards/curriculum_aware_reward_fn/mean": 0.1015625, "rewards/curriculum_aware_reward_fn/std": 0.1590231955051422, "step": 47 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1687.0, "completions/max_terminated_length": 1687.0, "completions/mean_length": 482.75225830078125, "completions/mean_terminated_length": 482.75225830078125, "completions/min_length": 179.0, "completions/min_terminated_length": 179.0, "epoch": 0.19804022692109335, "grad_norm": 0.17908784747123718, "kl": 0.026641845703125, "learning_rate": 1e-06, "loss": 0.0116, "num_tokens": 25115758.0, "reward": 1.094642996788025, "reward_std": 0.1125173270702362, "rewards/code_format_reward/mean": 0.9977678656578064, "rewards/code_format_reward/std": 0.047245558351278305, "rewards/curriculum_aware_reward_fn/mean": 0.09687499701976776, "rewards/curriculum_aware_reward_fn/std": 0.15676844120025635, "step": 48 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1909.0, "completions/max_terminated_length": 1909.0, "completions/mean_length": 488.2410888671875, "completions/mean_terminated_length": 488.2410888671875, "completions/min_length": 125.0, "completions/min_terminated_length": 125.0, "epoch": 0.20216606498194944, "grad_norm": 0.15744391083717346, "kl": 0.0242919921875, "learning_rate": 1e-06, "loss": 0.0197, "num_tokens": 25604424.0, "reward": 1.0954241752624512, "reward_std": 0.09485037624835968, "rewards/code_format_reward/mean": 0.9977678656578064, "rewards/code_format_reward/std": 0.047245558351278305, "rewards/curriculum_aware_reward_fn/mean": 0.09765625, "rewards/curriculum_aware_reward_fn/std": 0.16564589738845825, "step": 49 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2600.0, "completions/max_terminated_length": 2600.0, "completions/mean_length": 512.3683471679688, "completions/mean_terminated_length": 512.3683471679688, "completions/min_length": 143.0, "completions/min_terminated_length": 143.0, "epoch": 0.20629190304280556, "grad_norm": 0.15357136726379395, "kl": 0.0228729248046875, "learning_rate": 1e-06, "loss": 0.0092, "num_tokens": 26094421.0, "reward": 1.1697545051574707, "reward_std": 0.14999061822891235, "rewards/code_format_reward/mean": 0.9955357313156128, "rewards/code_format_reward/std": 0.06674052774906158, "rewards/curriculum_aware_reward_fn/mean": 0.17421874403953552, "rewards/curriculum_aware_reward_fn/std": 0.27402350306510925, "step": 50 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1301.0, "completions/max_terminated_length": 1301.0, "completions/mean_length": 488.3660888671875, "completions/mean_terminated_length": 488.3660888671875, "completions/min_length": 167.0, "completions/min_terminated_length": 167.0, "epoch": 0.21041774110366168, "grad_norm": 0.16300714015960693, "kl": 0.023529052734375, "learning_rate": 1e-06, "loss": 0.0173, "num_tokens": 26577109.0, "reward": 1.1524553298950195, "reward_std": 0.1449015736579895, "rewards/code_format_reward/mean": 0.9977678656578064, "rewards/code_format_reward/std": 0.047245558351278305, "rewards/curriculum_aware_reward_fn/mean": 0.15468750894069672, "rewards/curriculum_aware_reward_fn/std": 0.25778767466545105, "step": 51 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 1789.0, "completions/mean_length": 556.6495971679688, "completions/mean_terminated_length": 548.7315673828125, "completions/min_length": 212.0, "completions/min_terminated_length": 212.0, "epoch": 0.2145435791645178, "grad_norm": 0.1564698964357376, "kl": 0.0198516845703125, "learning_rate": 1e-06, "loss": 0.0106, "num_tokens": 27117541.0, "reward": 1.1220982074737549, "reward_std": 0.14631754159927368, "rewards/code_format_reward/mean": 0.9955357313156128, "rewards/code_format_reward/std": 0.06674052774906158, "rewards/curriculum_aware_reward_fn/mean": 0.12656249105930328, "rewards/curriculum_aware_reward_fn/std": 0.23956476151943207, "step": 52 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1865.0, "completions/max_terminated_length": 1865.0, "completions/mean_length": 522.607177734375, "completions/mean_terminated_length": 522.607177734375, "completions/min_length": 184.0, "completions/min_terminated_length": 184.0, "epoch": 0.2186694172253739, "grad_norm": 0.16431348025798798, "kl": 0.020294189453125, "learning_rate": 1e-06, "loss": 0.0147, "num_tokens": 27615485.0, "reward": 1.1806920766830444, "reward_std": 0.17966462671756744, "rewards/code_format_reward/mean": 0.9955357313156128, "rewards/code_format_reward/std": 0.06674052774906158, "rewards/curriculum_aware_reward_fn/mean": 0.18515624105930328, "rewards/curriculum_aware_reward_fn/std": 0.2865508198738098, "step": 53 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.004464285714285698, "completions/max_length": 4096.0, "completions/max_terminated_length": 1863.0, "completions/mean_length": 528.5580444335938, "completions/mean_terminated_length": 512.560546875, "completions/min_length": 116.0, "completions/min_terminated_length": 116.0, "epoch": 0.22279525528623, "grad_norm": 0.17439934611320496, "kl": 0.021240234375, "learning_rate": 1e-06, "loss": 0.0426, "num_tokens": 28105791.0, "reward": 1.127678632736206, "reward_std": 0.17251846194267273, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843916893005, "rewards/curriculum_aware_reward_fn/mean": 0.13437499105930328, "rewards/curriculum_aware_reward_fn/std": 0.23526187241077423, "step": 54 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1257.0, "completions/max_terminated_length": 1257.0, "completions/mean_length": 524.796875, "completions/mean_terminated_length": 524.796875, "completions/min_length": 206.0, "completions/min_terminated_length": 206.0, "epoch": 0.22692109334708613, "grad_norm": 0.17376989126205444, "kl": 0.022430419921875, "learning_rate": 1e-06, "loss": -0.001, "num_tokens": 28635222.0, "reward": 1.1344866752624512, "reward_std": 0.16489870846271515, "rewards/code_format_reward/mean": 0.9977678656578064, "rewards/code_format_reward/std": 0.047245558351278305, "rewards/curriculum_aware_reward_fn/mean": 0.1367187350988388, "rewards/curriculum_aware_reward_fn/std": 0.25031736493110657, "step": 55 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1768.0, "completions/max_terminated_length": 1768.0, "completions/mean_length": 536.8549194335938, "completions/mean_terminated_length": 536.8549194335938, "completions/min_length": 241.0, "completions/min_terminated_length": 241.0, "epoch": 0.23104693140794225, "grad_norm": 0.1522541344165802, "kl": 0.0203857421875, "learning_rate": 1e-06, "loss": 0.0224, "num_tokens": 29165477.0, "reward": 1.1148438453674316, "reward_std": 0.1370818018913269, "rewards/code_format_reward/mean": 1.0, "rewards/code_format_reward/std": 0.0, "rewards/curriculum_aware_reward_fn/mean": 0.11484374850988388, "rewards/curriculum_aware_reward_fn/std": 0.23221427202224731, "step": 56 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 1781.0, "completions/mean_length": 526.4486694335938, "completions/mean_terminated_length": 518.4630737304688, "completions/min_length": 170.0, "completions/min_terminated_length": 170.0, "epoch": 0.23517276946879834, "grad_norm": 0.17326720058918, "kl": 0.0211029052734375, "learning_rate": 1e-06, "loss": 0.0243, "num_tokens": 29679072.0, "reward": 1.2166296243667603, "reward_std": 0.17645685374736786, "rewards/code_format_reward/mean": 0.9955357313156128, "rewards/code_format_reward/std": 0.06674052774906158, "rewards/curriculum_aware_reward_fn/mean": 0.22109375894069672, "rewards/curriculum_aware_reward_fn/std": 0.3035515248775482, "step": 57 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1494.0, "completions/max_terminated_length": 1494.0, "completions/mean_length": 507.3013610839844, "completions/mean_terminated_length": 507.3013610839844, "completions/min_length": 177.0, "completions/min_terminated_length": 177.0, "epoch": 0.23929860752965446, "grad_norm": 0.16802732646465302, "kl": 0.01971435546875, "learning_rate": 1e-06, "loss": 0.0165, "num_tokens": 30167268.0, "reward": 1.1937501430511475, "reward_std": 0.17160995304584503, "rewards/code_format_reward/mean": 1.0, "rewards/code_format_reward/std": 0.0, "rewards/curriculum_aware_reward_fn/mean": 0.19374999403953552, "rewards/curriculum_aware_reward_fn/std": 0.2841939628124237, "step": 58 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1296.0, "completions/max_terminated_length": 1296.0, "completions/mean_length": 524.5982666015625, "completions/mean_terminated_length": 524.5982666015625, "completions/min_length": 161.0, "completions/min_terminated_length": 161.0, "epoch": 0.24342444559051057, "grad_norm": 0.14656943082809448, "kl": 0.02020263671875, "learning_rate": 1e-06, "loss": 0.0029, "num_tokens": 30672138.0, "reward": 1.1485490798950195, "reward_std": 0.12995967268943787, "rewards/code_format_reward/mean": 0.9977678656578064, "rewards/code_format_reward/std": 0.047245558351278305, "rewards/curriculum_aware_reward_fn/mean": 0.15078124403953552, "rewards/curriculum_aware_reward_fn/std": 0.2658287584781647, "step": 59 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 1761.0, "completions/mean_length": 518.841552734375, "completions/mean_terminated_length": 510.83892822265625, "completions/min_length": 209.0, "completions/min_terminated_length": 209.0, "epoch": 0.2475502836513667, "grad_norm": 0.14920854568481445, "kl": 0.0211639404296875, "learning_rate": 1e-06, "loss": 0.0328, "num_tokens": 31186070.0, "reward": 1.180580496788025, "reward_std": 0.15547817945480347, "rewards/code_format_reward/mean": 0.9977678656578064, "rewards/code_format_reward/std": 0.047245558351278305, "rewards/curriculum_aware_reward_fn/mean": 0.18281248211860657, "rewards/curriculum_aware_reward_fn/std": 0.2739129364490509, "step": 60 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 1185.0, "completions/mean_length": 529.25, "completions/mean_terminated_length": 521.2706909179688, "completions/min_length": 200.0, "completions/min_terminated_length": 200.0, "epoch": 0.2516761217122228, "grad_norm": 0.15654738247394562, "kl": 0.020477294921875, "learning_rate": 1e-06, "loss": 0.032, "num_tokens": 31683399.0, "reward": 1.1885044574737549, "reward_std": 0.17501546442508698, "rewards/code_format_reward/mean": 0.9955357313156128, "rewards/code_format_reward/std": 0.06674052774906158, "rewards/curriculum_aware_reward_fn/mean": 0.19296874105930328, "rewards/curriculum_aware_reward_fn/std": 0.2880752682685852, "step": 61 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 1066.0, "completions/mean_length": 527.6049194335938, "completions/mean_terminated_length": 519.6219482421875, "completions/min_length": 235.0, "completions/min_terminated_length": 235.0, "epoch": 0.2558019597730789, "grad_norm": 0.1481848806142807, "kl": 0.0185394287109375, "learning_rate": 1e-06, "loss": 0.0262, "num_tokens": 32203553.0, "reward": 1.139174222946167, "reward_std": 0.14721594750881195, "rewards/code_format_reward/mean": 0.9977678656578064, "rewards/code_format_reward/std": 0.047245558351278305, "rewards/curriculum_aware_reward_fn/mean": 0.14140625298023224, "rewards/curriculum_aware_reward_fn/std": 0.25424543023109436, "step": 62 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1218.0, "completions/max_terminated_length": 1218.0, "completions/mean_length": 516.9330444335938, "completions/mean_terminated_length": 516.9330444335938, "completions/min_length": 156.0, "completions/min_terminated_length": 156.0, "epoch": 0.259927797833935, "grad_norm": 0.18818791210651398, "kl": 0.03802490234375, "learning_rate": 1e-06, "loss": 0.0067, "num_tokens": 32697653.0, "reward": 1.1876118183135986, "reward_std": 0.18578869104385376, "rewards/code_format_reward/mean": 0.9977678656578064, "rewards/code_format_reward/std": 0.047245558351278305, "rewards/curriculum_aware_reward_fn/mean": 0.18984374403953552, "rewards/curriculum_aware_reward_fn/std": 0.30488669872283936, "step": 63 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1240.0, "completions/max_terminated_length": 1240.0, "completions/mean_length": 551.53125, "completions/mean_terminated_length": 551.53125, "completions/min_length": 184.0, "completions/min_terminated_length": 184.0, "epoch": 0.26405363589479114, "grad_norm": 0.16306030750274658, "kl": 0.0191497802734375, "learning_rate": 1e-06, "loss": 0.0033, "num_tokens": 33224620.0, "reward": 1.1351563930511475, "reward_std": 0.12161044031381607, "rewards/code_format_reward/mean": 1.0, "rewards/code_format_reward/std": 0.0, "rewards/curriculum_aware_reward_fn/mean": 0.13515624403953552, "rewards/curriculum_aware_reward_fn/std": 0.23886284232139587, "step": 64 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1202.0, "completions/max_terminated_length": 1202.0, "completions/mean_length": 526.8170166015625, "completions/mean_terminated_length": 526.8170166015625, "completions/min_length": 227.0, "completions/min_terminated_length": 227.0, "epoch": 0.26817947395564723, "grad_norm": 0.1532638669013977, "kl": 0.0195465087890625, "learning_rate": 1e-06, "loss": 0.0241, "num_tokens": 33730015.0, "reward": 1.1737724542617798, "reward_std": 0.17607219517230988, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843171834946, "rewards/curriculum_aware_reward_fn/mean": 0.18046875298023224, "rewards/curriculum_aware_reward_fn/std": 0.28667888045310974, "step": 65 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2010.0, "completions/max_terminated_length": 2010.0, "completions/mean_length": 521.3147583007812, "completions/mean_terminated_length": 521.3147583007812, "completions/min_length": 261.0, "completions/min_terminated_length": 261.0, "epoch": 0.2723053120165034, "grad_norm": 0.16237851977348328, "kl": 0.0205535888671875, "learning_rate": 1e-06, "loss": 0.0006, "num_tokens": 34233603.0, "reward": 1.192299246788025, "reward_std": 0.1378326714038849, "rewards/code_format_reward/mean": 0.9977678656578064, "rewards/code_format_reward/std": 0.047245558351278305, "rewards/curriculum_aware_reward_fn/mean": 0.19453123211860657, "rewards/curriculum_aware_reward_fn/std": 0.28702008724212646, "step": 66 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1043.0, "completions/max_terminated_length": 1043.0, "completions/mean_length": 503.1651916503906, "completions/mean_terminated_length": 503.1651916503906, "completions/min_length": 213.0, "completions/min_terminated_length": 213.0, "epoch": 0.27643115007735947, "grad_norm": 0.17482617497444153, "kl": 0.0332489013671875, "learning_rate": 1e-06, "loss": 0.0084, "num_tokens": 34716110.0, "reward": 1.1895090341567993, "reward_std": 0.1832786202430725, "rewards/code_format_reward/mean": 0.9910714030265808, "rewards/code_format_reward/std": 0.09417349100112915, "rewards/curriculum_aware_reward_fn/mean": 0.19843748211860657, "rewards/curriculum_aware_reward_fn/std": 0.3167010545730591, "step": 67 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1179.0, "completions/max_terminated_length": 1179.0, "completions/mean_length": 523.0178833007812, "completions/mean_terminated_length": 523.0178833007812, "completions/min_length": 223.0, "completions/min_terminated_length": 223.0, "epoch": 0.28055698813821556, "grad_norm": 0.15600912272930145, "kl": 0.0208740234375, "learning_rate": 1e-06, "loss": 0.0081, "num_tokens": 35214700.0, "reward": 1.1719865798950195, "reward_std": 0.14628802239894867, "rewards/code_format_reward/mean": 0.9977678656578064, "rewards/code_format_reward/std": 0.047245558351278305, "rewards/curriculum_aware_reward_fn/mean": 0.17421875894069672, "rewards/curriculum_aware_reward_fn/std": 0.27402347326278687, "step": 68 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1340.0, "completions/max_terminated_length": 1340.0, "completions/mean_length": 518.732177734375, "completions/mean_terminated_length": 518.732177734375, "completions/min_length": 231.0, "completions/min_terminated_length": 231.0, "epoch": 0.2846828261990717, "grad_norm": 0.15977604687213898, "kl": 0.0284271240234375, "learning_rate": 1e-06, "loss": 0.013, "num_tokens": 35705084.0, "reward": 1.1867188215255737, "reward_std": 0.13772818446159363, "rewards/code_format_reward/mean": 1.0, "rewards/code_format_reward/std": 0.0, "rewards/curriculum_aware_reward_fn/mean": 0.18671874701976776, "rewards/curriculum_aware_reward_fn/std": 0.2787334620952606, "step": 69 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 1527.0, "completions/mean_length": 540.4330444335938, "completions/mean_terminated_length": 532.478759765625, "completions/min_length": 206.0, "completions/min_terminated_length": 206.0, "epoch": 0.2888086642599278, "grad_norm": 0.16505984961986542, "kl": 0.0190582275390625, "learning_rate": 1e-06, "loss": 0.0162, "num_tokens": 36202817.0, "reward": 1.1292412281036377, "reward_std": 0.17226025462150574, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843171834946, "rewards/curriculum_aware_reward_fn/mean": 0.13593749701976776, "rewards/curriculum_aware_reward_fn/std": 0.24127459526062012, "step": 70 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 1484.0, "completions/mean_length": 526.5892944335938, "completions/mean_terminated_length": 518.60400390625, "completions/min_length": 242.0, "completions/min_terminated_length": 242.0, "epoch": 0.2929345023207839, "grad_norm": 0.15790706872940063, "kl": 0.0190887451171875, "learning_rate": 1e-06, "loss": 0.0309, "num_tokens": 36712401.0, "reward": 1.192522406578064, "reward_std": 0.1598421186208725, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843916893005, "rewards/curriculum_aware_reward_fn/mean": 0.19921875, "rewards/curriculum_aware_reward_fn/std": 0.29234156012535095, "step": 71 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1069.0, "completions/max_terminated_length": 1069.0, "completions/mean_length": 550.3192138671875, "completions/mean_terminated_length": 550.3192138671875, "completions/min_length": 257.0, "completions/min_terminated_length": 257.0, "epoch": 0.29706034038164003, "grad_norm": 0.16435663402080536, "kl": 0.0229034423828125, "learning_rate": 1e-06, "loss": 0.0059, "num_tokens": 37239399.0, "reward": 1.1632813215255737, "reward_std": 0.18066853284835815, "rewards/code_format_reward/mean": 0.984375, "rewards/code_format_reward/std": 0.12415824085474014, "rewards/curriculum_aware_reward_fn/mean": 0.17890623211860657, "rewards/curriculum_aware_reward_fn/std": 0.2689492404460907, "step": 72 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 1790.0, "completions/mean_length": 529.9933471679688, "completions/mean_terminated_length": 522.0156860351562, "completions/min_length": 233.0, "completions/min_terminated_length": 233.0, "epoch": 0.3011861784424961, "grad_norm": 0.15893499553203583, "kl": 0.0205230712890625, "learning_rate": 1e-06, "loss": 0.0333, "num_tokens": 37742087.0, "reward": 1.1675224304199219, "reward_std": 0.1739296019077301, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843171834946, "rewards/curriculum_aware_reward_fn/mean": 0.17421875894069672, "rewards/curriculum_aware_reward_fn/std": 0.2857727110385895, "step": 73 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 1741.0, "completions/mean_length": 580.841552734375, "completions/mean_terminated_length": 572.9776611328125, "completions/min_length": 214.0, "completions/min_terminated_length": 214.0, "epoch": 0.30531201650335227, "grad_norm": 0.15558117628097534, "kl": 0.018707275390625, "learning_rate": 1e-06, "loss": 0.0291, "num_tokens": 38295107.0, "reward": 1.1248884201049805, "reward_std": 0.167986199259758, "rewards/code_format_reward/mean": 0.9866071343421936, "rewards/code_format_reward/std": 0.11507844924926758, "rewards/curriculum_aware_reward_fn/mean": 0.13828125596046448, "rewards/curriculum_aware_reward_fn/std": 0.25164324045181274, "step": 74 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1512.0, "completions/max_terminated_length": 1512.0, "completions/mean_length": 547.21875, "completions/mean_terminated_length": 547.21875, "completions/min_length": 277.0, "completions/min_terminated_length": 277.0, "epoch": 0.30943785456420836, "grad_norm": 0.166014164686203, "kl": 0.021240234375, "learning_rate": 1e-06, "loss": 0.0013, "num_tokens": 38806604.0, "reward": 1.1474331617355347, "reward_std": 0.16723909974098206, "rewards/code_format_reward/mean": 0.9888392686843872, "rewards/code_format_reward/std": 0.10517053306102753, "rewards/curriculum_aware_reward_fn/mean": 0.15859375894069672, "rewards/curriculum_aware_reward_fn/std": 0.2814329266548157, "step": 75 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 1560.0, "completions/mean_length": 574.6205444335938, "completions/mean_terminated_length": 566.7427368164062, "completions/min_length": 278.0, "completions/min_terminated_length": 278.0, "epoch": 0.31356369262506445, "grad_norm": 0.22545717656612396, "kl": 0.02032470703125, "learning_rate": 1e-06, "loss": 0.0227, "num_tokens": 39349011.0, "reward": 1.1671875715255737, "reward_std": 0.18748165667057037, "rewards/code_format_reward/mean": 0.984375, "rewards/code_format_reward/std": 0.12415824085474014, "rewards/curriculum_aware_reward_fn/mean": 0.18281248211860657, "rewards/curriculum_aware_reward_fn/std": 0.28180328011512756, "step": 76 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 2560.0, "completions/mean_length": 604.1138916015625, "completions/mean_terminated_length": 596.302001953125, "completions/min_length": 266.0, "completions/min_terminated_length": 266.0, "epoch": 0.3176895306859206, "grad_norm": 0.14813967049121857, "kl": 0.0190582275390625, "learning_rate": 1e-06, "loss": 0.0173, "num_tokens": 39898968.0, "reward": 1.1960939168930054, "reward_std": 0.1797778308391571, "rewards/code_format_reward/mean": 0.984375, "rewards/code_format_reward/std": 0.12415824085474014, "rewards/curriculum_aware_reward_fn/mean": 0.21171875298023224, "rewards/curriculum_aware_reward_fn/std": 0.31974881887435913, "step": 77 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 1447.0, "completions/mean_length": 583.9866333007812, "completions/mean_terminated_length": 576.1297607421875, "completions/min_length": 235.0, "completions/min_terminated_length": 235.0, "epoch": 0.3218153687467767, "grad_norm": 0.16099686920642853, "kl": 0.0206298828125, "learning_rate": 1e-06, "loss": 0.0246, "num_tokens": 40436142.0, "reward": 1.1539063453674316, "reward_std": 0.17733997106552124, "rewards/code_format_reward/mean": 0.984375, "rewards/code_format_reward/std": 0.12415824085474014, "rewards/curriculum_aware_reward_fn/mean": 0.16953124105930328, "rewards/curriculum_aware_reward_fn/std": 0.2699390947818756, "step": 78 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1446.0, "completions/max_terminated_length": 1446.0, "completions/mean_length": 571.1473388671875, "completions/mean_terminated_length": 571.1473388671875, "completions/min_length": 235.0, "completions/min_terminated_length": 235.0, "epoch": 0.3259412068076328, "grad_norm": 0.15308991074562073, "kl": 0.0211029052734375, "learning_rate": 1e-06, "loss": 0.0054, "num_tokens": 40963301.0, "reward": 1.1388393640518188, "reward_std": 0.14730839431285858, "rewards/code_format_reward/mean": 0.9888392686843872, "rewards/code_format_reward/std": 0.10517053306102753, "rewards/curriculum_aware_reward_fn/mean": 0.14999999105930328, "rewards/curriculum_aware_reward_fn/std": 0.26368576288223267, "step": 79 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1335.0, "completions/max_terminated_length": 1335.0, "completions/mean_length": 568.2232666015625, "completions/mean_terminated_length": 568.2232666015625, "completions/min_length": 199.0, "completions/min_terminated_length": 199.0, "epoch": 0.3300670448684889, "grad_norm": 0.16148877143859863, "kl": 0.021453857421875, "learning_rate": 1e-06, "loss": 0.0083, "num_tokens": 41480155.0, "reward": 1.14453125, "reward_std": 0.18031297624111176, "rewards/code_format_reward/mean": 0.984375, "rewards/code_format_reward/std": 0.12415824085474014, "rewards/curriculum_aware_reward_fn/mean": 0.16015625, "rewards/curriculum_aware_reward_fn/std": 0.26958534121513367, "step": 80 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1757.0, "completions/max_terminated_length": 1757.0, "completions/mean_length": 559.8080444335938, "completions/mean_terminated_length": 559.8080444335938, "completions/min_length": 218.0, "completions/min_terminated_length": 218.0, "epoch": 0.334192882929345, "grad_norm": 0.13826853036880493, "kl": 0.022613525390625, "learning_rate": 1e-06, "loss": 0.0166, "num_tokens": 42002353.0, "reward": 1.2103794813156128, "reward_std": 0.14001040160655975, "rewards/code_format_reward/mean": 0.9955357313156128, "rewards/code_format_reward/std": 0.06674052774906158, "rewards/curriculum_aware_reward_fn/mean": 0.21484375, "rewards/curriculum_aware_reward_fn/std": 0.3150540292263031, "step": 81 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1970.0, "completions/max_terminated_length": 1970.0, "completions/mean_length": 583.6875, "completions/mean_terminated_length": 583.6875, "completions/min_length": 235.0, "completions/min_terminated_length": 235.0, "epoch": 0.33831872099020116, "grad_norm": 0.14655961096286774, "kl": 0.01815032958984375, "learning_rate": 1e-06, "loss": 0.0147, "num_tokens": 42532169.0, "reward": 1.16015625, "reward_std": 0.16172908246517181, "rewards/code_format_reward/mean": 0.984375, "rewards/code_format_reward/std": 0.12415824085474014, "rewards/curriculum_aware_reward_fn/mean": 0.17578125, "rewards/curriculum_aware_reward_fn/std": 0.2848121225833893, "step": 82 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2949.0, "completions/max_terminated_length": 2949.0, "completions/mean_length": 617.0357666015625, "completions/mean_terminated_length": 617.0357666015625, "completions/min_length": 234.0, "completions/min_terminated_length": 234.0, "epoch": 0.34244455905105725, "grad_norm": 0.16233304142951965, "kl": 0.01851654052734375, "learning_rate": 1e-06, "loss": 0.0184, "num_tokens": 43100585.0, "reward": 1.161384105682373, "reward_std": 0.19341857731342316, "rewards/code_format_reward/mean": 0.9910714030265808, "rewards/code_format_reward/std": 0.09417349100112915, "rewards/curriculum_aware_reward_fn/mean": 0.17031250894069672, "rewards/curriculum_aware_reward_fn/std": 0.26689091324806213, "step": 83 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2658.0, "completions/max_terminated_length": 2658.0, "completions/mean_length": 649.1205444335938, "completions/mean_terminated_length": 649.1205444335938, "completions/min_length": 255.0, "completions/min_terminated_length": 255.0, "epoch": 0.34657039711191334, "grad_norm": 0.14131295680999756, "kl": 0.019439697265625, "learning_rate": 1e-06, "loss": 0.0151, "num_tokens": 43674324.0, "reward": 1.111830472946167, "reward_std": 0.14097417891025543, "rewards/code_format_reward/mean": 0.9821428656578064, "rewards/code_format_reward/std": 0.13258016109466553, "rewards/curriculum_aware_reward_fn/mean": 0.12968750298023224, "rewards/curriculum_aware_reward_fn/std": 0.24244815111160278, "step": 84 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1551.0, "completions/max_terminated_length": 1551.0, "completions/mean_length": 611.125, "completions/mean_terminated_length": 611.125, "completions/min_length": 253.0, "completions/min_terminated_length": 253.0, "epoch": 0.3506962351727695, "grad_norm": 0.16218431293964386, "kl": 0.0191497802734375, "learning_rate": 1e-06, "loss": 0.0098, "num_tokens": 44228836.0, "reward": 1.1375001668930054, "reward_std": 0.18307891488075256, "rewards/code_format_reward/mean": 0.984375, "rewards/code_format_reward/std": 0.12415824085474014, "rewards/curriculum_aware_reward_fn/mean": 0.15312500298023224, "rewards/curriculum_aware_reward_fn/std": 0.25119680166244507, "step": 85 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1560.0, "completions/max_terminated_length": 1560.0, "completions/mean_length": 588.6785888671875, "completions/mean_terminated_length": 588.6785888671875, "completions/min_length": 256.0, "completions/min_terminated_length": 256.0, "epoch": 0.3548220732336256, "grad_norm": 0.1462087631225586, "kl": 0.0193023681640625, "learning_rate": 1e-06, "loss": 0.0094, "num_tokens": 44750108.0, "reward": 1.1293528079986572, "reward_std": 0.157430499792099, "rewards/code_format_reward/mean": 0.9910714030265808, "rewards/code_format_reward/std": 0.09417349845170975, "rewards/curriculum_aware_reward_fn/mean": 0.13828124105930328, "rewards/curriculum_aware_reward_fn/std": 0.2461378425359726, "step": 86 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1634.0, "completions/max_terminated_length": 1634.0, "completions/mean_length": 630.6674194335938, "completions/mean_terminated_length": 630.6674194335938, "completions/min_length": 224.0, "completions/min_terminated_length": 224.0, "epoch": 0.35894791129448167, "grad_norm": 0.14705541729927063, "kl": 0.0197906494140625, "learning_rate": 1e-06, "loss": 0.0038, "num_tokens": 45317001.0, "reward": 1.1686384677886963, "reward_std": 0.1696617752313614, "rewards/code_format_reward/mean": 0.9866071343421936, "rewards/code_format_reward/std": 0.11507843434810638, "rewards/curriculum_aware_reward_fn/mean": 0.18203125894069672, "rewards/curriculum_aware_reward_fn/std": 0.280849814414978, "step": 87 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 3749.0, "completions/max_terminated_length": 3749.0, "completions/mean_length": 625.779052734375, "completions/mean_terminated_length": 625.779052734375, "completions/min_length": 251.0, "completions/min_terminated_length": 251.0, "epoch": 0.3630737493553378, "grad_norm": 0.15462744235992432, "kl": 0.020904541015625, "learning_rate": 1e-06, "loss": 0.0114, "num_tokens": 45873066.0, "reward": 1.1773439645767212, "reward_std": 0.16795016825199127, "rewards/code_format_reward/mean": 0.984375, "rewards/code_format_reward/std": 0.12415824085474014, "rewards/curriculum_aware_reward_fn/mean": 0.19296874105930328, "rewards/curriculum_aware_reward_fn/std": 0.284244567155838, "step": 88 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 3013.0, "completions/max_terminated_length": 3013.0, "completions/mean_length": 638.1451416015625, "completions/mean_terminated_length": 638.1451416015625, "completions/min_length": 201.0, "completions/min_terminated_length": 201.0, "epoch": 0.3671995874161939, "grad_norm": 0.13075420260429382, "kl": 0.0183868408203125, "learning_rate": 1e-06, "loss": 0.009, "num_tokens": 46440831.0, "reward": 1.1301339864730835, "reward_std": 0.1329827904701233, "rewards/code_format_reward/mean": 0.9910714030265808, "rewards/code_format_reward/std": 0.09417349100112915, "rewards/curriculum_aware_reward_fn/mean": 0.13906249403953552, "rewards/curriculum_aware_reward_fn/std": 0.2528424561023712, "step": 89 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1418.0, "completions/max_terminated_length": 1418.0, "completions/mean_length": 626.5535888671875, "completions/mean_terminated_length": 626.5535888671875, "completions/min_length": 251.0, "completions/min_terminated_length": 251.0, "epoch": 0.37132542547705005, "grad_norm": 0.1601579189300537, "kl": 0.0171966552734375, "learning_rate": 1e-06, "loss": 0.0114, "num_tokens": 47001058.0, "reward": 1.172991156578064, "reward_std": 0.18612328171730042, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843916893005, "rewards/curriculum_aware_reward_fn/mean": 0.1796875, "rewards/curriculum_aware_reward_fn/std": 0.2658621072769165, "step": 90 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 3307.0, "completions/max_terminated_length": 3307.0, "completions/mean_length": 626.7611694335938, "completions/mean_terminated_length": 626.7611694335938, "completions/min_length": 231.0, "completions/min_terminated_length": 231.0, "epoch": 0.37545126353790614, "grad_norm": 0.1480480581521988, "kl": 0.022064208984375, "learning_rate": 1e-06, "loss": 0.0186, "num_tokens": 47560027.0, "reward": 1.142076015472412, "reward_std": 0.16457779705524445, "rewards/code_format_reward/mean": 0.9866071343421936, "rewards/code_format_reward/std": 0.11507843434810638, "rewards/curriculum_aware_reward_fn/mean": 0.15546873211860657, "rewards/curriculum_aware_reward_fn/std": 0.266215056180954, "step": 91 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1551.0, "completions/max_terminated_length": 1551.0, "completions/mean_length": 634.09375, "completions/mean_terminated_length": 634.09375, "completions/min_length": 276.0, "completions/min_terminated_length": 276.0, "epoch": 0.37957710159876223, "grad_norm": 0.14755026996135712, "kl": 0.0177001953125, "learning_rate": 1e-06, "loss": 0.0008, "num_tokens": 48102765.0, "reward": 1.138169765472412, "reward_std": 0.1581386774778366, "rewards/code_format_reward/mean": 0.9866071343421936, "rewards/code_format_reward/std": 0.11507843434810638, "rewards/curriculum_aware_reward_fn/mean": 0.15156248211860657, "rewards/curriculum_aware_reward_fn/std": 0.27897584438323975, "step": 92 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2012.0, "completions/max_terminated_length": 2012.0, "completions/mean_length": 624.1451416015625, "completions/mean_terminated_length": 624.1451416015625, "completions/min_length": 244.0, "completions/min_terminated_length": 244.0, "epoch": 0.3837029396596184, "grad_norm": 0.1583220362663269, "kl": 0.0184783935546875, "learning_rate": 1e-06, "loss": 0.013, "num_tokens": 48655851.0, "reward": 1.1885045766830444, "reward_std": 0.22364631295204163, "rewards/code_format_reward/mean": 0.9799107313156128, "rewards/code_format_reward/std": 0.14046260714530945, "rewards/curriculum_aware_reward_fn/mean": 0.20859374105930328, "rewards/curriculum_aware_reward_fn/std": 0.3033902943134308, "step": 93 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 3739.0, "completions/max_terminated_length": 3739.0, "completions/mean_length": 642.0045166015625, "completions/mean_terminated_length": 642.0045166015625, "completions/min_length": 210.0, "completions/min_terminated_length": 210.0, "epoch": 0.38782877772047447, "grad_norm": 0.14486397802829742, "kl": 0.0182037353515625, "learning_rate": 1e-06, "loss": 0.0174, "num_tokens": 49208979.0, "reward": 1.167745590209961, "reward_std": 0.1853606402873993, "rewards/code_format_reward/mean": 0.9888392686843872, "rewards/code_format_reward/std": 0.10517053306102753, "rewards/curriculum_aware_reward_fn/mean": 0.17890624701976776, "rewards/curriculum_aware_reward_fn/std": 0.28670451045036316, "step": 94 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1520.0, "completions/max_terminated_length": 1520.0, "completions/mean_length": 629.0670166015625, "completions/mean_terminated_length": 629.0670166015625, "completions/min_length": 275.0, "completions/min_terminated_length": 275.0, "epoch": 0.39195461578133056, "grad_norm": 0.14910699427127838, "kl": 0.0187225341796875, "learning_rate": 1e-06, "loss": 0.0079, "num_tokens": 49766531.0, "reward": 1.1893974542617798, "reward_std": 0.1598261594772339, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843916893005, "rewards/curriculum_aware_reward_fn/mean": 0.19609375298023224, "rewards/curriculum_aware_reward_fn/std": 0.3036160171031952, "step": 95 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1813.0, "completions/max_terminated_length": 1813.0, "completions/mean_length": 624.9553833007812, "completions/mean_terminated_length": 624.9553833007812, "completions/min_length": 205.0, "completions/min_terminated_length": 205.0, "epoch": 0.3960804538421867, "grad_norm": 0.14866606891155243, "kl": 0.0218658447265625, "learning_rate": 1e-06, "loss": 0.0108, "num_tokens": 50317740.0, "reward": 1.1503348350524902, "reward_std": 0.17856933176517487, "rewards/code_format_reward/mean": 0.9776785969734192, "rewards/code_format_reward/std": 0.1478918492794037, "rewards/curriculum_aware_reward_fn/mean": 0.17265625298023224, "rewards/curriculum_aware_reward_fn/std": 0.2769986689090729, "step": 96 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1349.0, "completions/max_terminated_length": 1349.0, "completions/mean_length": 639.8147583007812, "completions/mean_terminated_length": 639.8147583007812, "completions/min_length": 225.0, "completions/min_terminated_length": 225.0, "epoch": 0.4002062919030428, "grad_norm": 0.15107698738574982, "kl": 0.0175933837890625, "learning_rate": 1e-06, "loss": 0.0001, "num_tokens": 50870356.0, "reward": 1.1937501430511475, "reward_std": 0.18680307269096375, "rewards/code_format_reward/mean": 0.984375, "rewards/code_format_reward/std": 0.12415824085474014, "rewards/curriculum_aware_reward_fn/mean": 0.20937499403953552, "rewards/curriculum_aware_reward_fn/std": 0.318292498588562, "step": 97 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1881.0, "completions/max_terminated_length": 1881.0, "completions/mean_length": 654.763427734375, "completions/mean_terminated_length": 654.763427734375, "completions/min_length": 159.0, "completions/min_terminated_length": 159.0, "epoch": 0.4043321299638989, "grad_norm": 0.1600561887025833, "kl": 0.0186920166015625, "learning_rate": 1e-06, "loss": 0.0078, "num_tokens": 51439364.0, "reward": 1.195312738418579, "reward_std": 0.20213094353675842, "rewards/code_format_reward/mean": 0.9866071343421936, "rewards/code_format_reward/std": 0.11507843434810638, "rewards/curriculum_aware_reward_fn/mean": 0.2087053507566452, "rewards/curriculum_aware_reward_fn/std": 0.3229687809944153, "step": 98 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1246.0, "completions/max_terminated_length": 1246.0, "completions/mean_length": 602.3616333007812, "completions/mean_terminated_length": 602.3616333007812, "completions/min_length": 267.0, "completions/min_terminated_length": 267.0, "epoch": 0.40845796802475504, "grad_norm": 0.13607917726039886, "kl": 0.0199737548828125, "learning_rate": 1e-06, "loss": 0.0101, "num_tokens": 51979436.0, "reward": 1.1448662281036377, "reward_std": 0.1558069884777069, "rewards/code_format_reward/mean": 0.9955357313156128, "rewards/code_format_reward/std": 0.06674052774906158, "rewards/curriculum_aware_reward_fn/mean": 0.14933036267757416, "rewards/curriculum_aware_reward_fn/std": 0.29378268122673035, "step": 99 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1603.0, "completions/max_terminated_length": 1603.0, "completions/mean_length": 651.875, "completions/mean_terminated_length": 651.875, "completions/min_length": 247.0, "completions/min_terminated_length": 247.0, "epoch": 0.4125838060856111, "grad_norm": 0.12900599837303162, "kl": 0.018524169921875, "learning_rate": 1e-06, "loss": 0.0221, "num_tokens": 52536111.0, "reward": 1.1502232551574707, "reward_std": 0.16727718710899353, "rewards/code_format_reward/mean": 0.9866071343421936, "rewards/code_format_reward/std": 0.11507844179868698, "rewards/curriculum_aware_reward_fn/mean": 0.16361607611179352, "rewards/curriculum_aware_reward_fn/std": 0.31127917766571045, "step": 100 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1762.0, "completions/max_terminated_length": 1762.0, "completions/mean_length": 655.2522583007812, "completions/mean_terminated_length": 655.2522583007812, "completions/min_length": 267.0, "completions/min_terminated_length": 267.0, "epoch": 0.4167096441464673, "grad_norm": 0.14474211633205414, "kl": 0.01849365234375, "learning_rate": 1e-06, "loss": 0.0099, "num_tokens": 53112674.0, "reward": 1.182142972946167, "reward_std": 0.15763171017169952, "rewards/code_format_reward/mean": 0.9977678656578064, "rewards/code_format_reward/std": 0.047245558351278305, "rewards/curriculum_aware_reward_fn/mean": 0.18437501788139343, "rewards/curriculum_aware_reward_fn/std": 0.3020090162754059, "step": 101 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2849.0, "completions/max_terminated_length": 2849.0, "completions/mean_length": 648.9866333007812, "completions/mean_terminated_length": 648.9866333007812, "completions/min_length": 264.0, "completions/min_terminated_length": 264.0, "epoch": 0.42083548220732336, "grad_norm": 0.1270856410264969, "kl": 0.0184173583984375, "learning_rate": 1e-06, "loss": 0.007, "num_tokens": 53688768.0, "reward": 1.1958706378936768, "reward_std": 0.17443805932998657, "rewards/code_format_reward/mean": 0.9866071343421936, "rewards/code_format_reward/std": 0.11507843434810638, "rewards/curriculum_aware_reward_fn/mean": 0.2092633992433548, "rewards/curriculum_aware_reward_fn/std": 0.33975598216056824, "step": 102 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2069.0, "completions/max_terminated_length": 2069.0, "completions/mean_length": 667.84375, "completions/mean_terminated_length": 667.84375, "completions/min_length": 282.0, "completions/min_terminated_length": 282.0, "epoch": 0.42496132026817945, "grad_norm": 0.1485375016927719, "kl": 0.016845703125, "learning_rate": 1e-06, "loss": 0.0292, "num_tokens": 54268396.0, "reward": 1.165513515472412, "reward_std": 0.1980009227991104, "rewards/code_format_reward/mean": 0.9866071343421936, "rewards/code_format_reward/std": 0.11507843434810638, "rewards/curriculum_aware_reward_fn/mean": 0.17890623211860657, "rewards/curriculum_aware_reward_fn/std": 0.30522361397743225, "step": 103 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1834.0, "completions/max_terminated_length": 1834.0, "completions/mean_length": 684.7076416015625, "completions/mean_terminated_length": 684.7076416015625, "completions/min_length": 304.0, "completions/min_terminated_length": 304.0, "epoch": 0.4290871583290356, "grad_norm": 0.1125236228108406, "kl": 0.01686859130859375, "learning_rate": 1e-06, "loss": 0.032, "num_tokens": 54831128.0, "reward": 1.0866073369979858, "reward_std": 0.1270635575056076, "rewards/code_format_reward/mean": 0.9866071343421936, "rewards/code_format_reward/std": 0.11507843434810638, "rewards/curriculum_aware_reward_fn/mean": 0.10000000149011612, "rewards/curriculum_aware_reward_fn/std": 0.2362789362668991, "step": 104 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1543.0, "completions/max_terminated_length": 1543.0, "completions/mean_length": 596.2611694335938, "completions/mean_terminated_length": 596.2611694335938, "completions/min_length": 278.0, "completions/min_terminated_length": 278.0, "epoch": 0.4332129963898917, "grad_norm": 0.15816938877105713, "kl": 0.02044677734375, "learning_rate": 1e-06, "loss": 0.0001, "num_tokens": 55358331.0, "reward": 1.239174246788025, "reward_std": 0.1988132745027542, "rewards/code_format_reward/mean": 0.9888392686843872, "rewards/code_format_reward/std": 0.10517053306102753, "rewards/curriculum_aware_reward_fn/mean": 0.25033482909202576, "rewards/curriculum_aware_reward_fn/std": 0.34741759300231934, "step": 105 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2732.0, "completions/max_terminated_length": 2732.0, "completions/mean_length": 644.7678833007812, "completions/mean_terminated_length": 644.7678833007812, "completions/min_length": 259.0, "completions/min_terminated_length": 259.0, "epoch": 0.4373388344507478, "grad_norm": 0.14656005799770355, "kl": 0.019317626953125, "learning_rate": 1e-06, "loss": 0.0038, "num_tokens": 55921223.0, "reward": 1.1698660850524902, "reward_std": 0.186209574341774, "rewards/code_format_reward/mean": 0.9955357313156128, "rewards/code_format_reward/std": 0.06674052774906158, "rewards/curriculum_aware_reward_fn/mean": 0.17433036863803864, "rewards/curriculum_aware_reward_fn/std": 0.3043120205402374, "step": 106 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2028.0, "completions/max_terminated_length": 2028.0, "completions/mean_length": 628.6785888671875, "completions/mean_terminated_length": 628.6785888671875, "completions/min_length": 269.0, "completions/min_terminated_length": 269.0, "epoch": 0.44146467251160393, "grad_norm": 0.15770718455314636, "kl": 0.02008056640625, "learning_rate": 1e-06, "loss": 0.0026, "num_tokens": 56455676.0, "reward": 1.256361722946167, "reward_std": 0.17305786907672882, "rewards/code_format_reward/mean": 1.0, "rewards/code_format_reward/std": 0.0, "rewards/curriculum_aware_reward_fn/mean": 0.25636163353919983, "rewards/curriculum_aware_reward_fn/std": 0.33519959449768066, "step": 107 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2329.0, "completions/max_terminated_length": 2329.0, "completions/mean_length": 643.875, "completions/mean_terminated_length": 643.875, "completions/min_length": 295.0, "completions/min_terminated_length": 295.0, "epoch": 0.44559051057246, "grad_norm": 0.1529017835855484, "kl": 0.02215576171875, "learning_rate": 1e-06, "loss": 0.0072, "num_tokens": 57012260.0, "reward": 1.2303571701049805, "reward_std": 0.21905755996704102, "rewards/code_format_reward/mean": 0.9866071343421936, "rewards/code_format_reward/std": 0.11507844179868698, "rewards/curriculum_aware_reward_fn/mean": 0.24374999105930328, "rewards/curriculum_aware_reward_fn/std": 0.3518964350223541, "step": 108 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2024.0, "completions/max_terminated_length": 2024.0, "completions/mean_length": 634.9933471679688, "completions/mean_terminated_length": 634.9933471679688, "completions/min_length": 285.0, "completions/min_terminated_length": 285.0, "epoch": 0.44971634863331617, "grad_norm": 0.12553778290748596, "kl": 0.0210418701171875, "learning_rate": 1e-06, "loss": 0.0097, "num_tokens": 57564116.0, "reward": 1.2456474304199219, "reward_std": 0.14504222571849823, "rewards/code_format_reward/mean": 0.9955357313156128, "rewards/code_format_reward/std": 0.06674052774906158, "rewards/curriculum_aware_reward_fn/mean": 0.2501116096973419, "rewards/curriculum_aware_reward_fn/std": 0.3558790385723114, "step": 109 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2150.0, "completions/max_terminated_length": 2150.0, "completions/mean_length": 682.9598388671875, "completions/mean_terminated_length": 682.9598388671875, "completions/min_length": 291.0, "completions/min_terminated_length": 291.0, "epoch": 0.45384218669417226, "grad_norm": 0.14662572741508484, "kl": 0.0195159912109375, "learning_rate": 1e-06, "loss": 0.0084, "num_tokens": 58142102.0, "reward": 1.1831474304199219, "reward_std": 0.19062189757823944, "rewards/code_format_reward/mean": 0.9866071343421936, "rewards/code_format_reward/std": 0.11507843434810638, "rewards/curriculum_aware_reward_fn/mean": 0.19654019176959991, "rewards/curriculum_aware_reward_fn/std": 0.3277512788772583, "step": 110 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1743.0, "completions/max_terminated_length": 1743.0, "completions/mean_length": 626.8817138671875, "completions/mean_terminated_length": 626.8817138671875, "completions/min_length": 250.0, "completions/min_terminated_length": 250.0, "epoch": 0.45796802475502835, "grad_norm": 0.13702163100242615, "kl": 0.0236358642578125, "learning_rate": 1e-06, "loss": 0.0116, "num_tokens": 58668393.0, "reward": 1.2010046243667603, "reward_std": 0.13696451485157013, "rewards/code_format_reward/mean": 0.9955357313156128, "rewards/code_format_reward/std": 0.06674052774906158, "rewards/curriculum_aware_reward_fn/mean": 0.20546875894069672, "rewards/curriculum_aware_reward_fn/std": 0.3315751254558563, "step": 111 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2658.0, "completions/max_terminated_length": 2658.0, "completions/mean_length": 635.4241333007812, "completions/mean_terminated_length": 635.4241333007812, "completions/min_length": 171.0, "completions/min_terminated_length": 171.0, "epoch": 0.4620938628158845, "grad_norm": 0.13016532361507416, "kl": 0.022674560546875, "learning_rate": 1e-06, "loss": 0.0074, "num_tokens": 59234974.0, "reward": 1.1763393878936768, "reward_std": 0.13711431622505188, "rewards/code_format_reward/mean": 0.9977678656578064, "rewards/code_format_reward/std": 0.047245558351278305, "rewards/curriculum_aware_reward_fn/mean": 0.1785714328289032, "rewards/curriculum_aware_reward_fn/std": 0.2989889085292816, "step": 112 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1847.0, "completions/max_terminated_length": 1847.0, "completions/mean_length": 678.716552734375, "completions/mean_terminated_length": 678.716552734375, "completions/min_length": 283.0, "completions/min_terminated_length": 283.0, "epoch": 0.4662197008767406, "grad_norm": 0.13799616694450378, "kl": 0.01947021484375, "learning_rate": 1e-06, "loss": -0.0054, "num_tokens": 59824226.0, "reward": 1.1936384439468384, "reward_std": 0.16037216782569885, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843916893005, "rewards/curriculum_aware_reward_fn/mean": 0.2003348171710968, "rewards/curriculum_aware_reward_fn/std": 0.3211154043674469, "step": 113 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1965.0, "completions/max_terminated_length": 1965.0, "completions/mean_length": 648.609375, "completions/mean_terminated_length": 648.609375, "completions/min_length": 235.0, "completions/min_terminated_length": 235.0, "epoch": 0.4703455389375967, "grad_norm": 0.13693882524967194, "kl": 0.019775390625, "learning_rate": 1e-06, "loss": -0.0081, "num_tokens": 60366992.0, "reward": 1.186272382736206, "reward_std": 0.15267953276634216, "rewards/code_format_reward/mean": 0.9977678656578064, "rewards/code_format_reward/std": 0.047245558351278305, "rewards/curriculum_aware_reward_fn/mean": 0.18850445747375488, "rewards/curriculum_aware_reward_fn/std": 0.31868821382522583, "step": 114 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2517.0, "completions/max_terminated_length": 2517.0, "completions/mean_length": 659.3951416015625, "completions/mean_terminated_length": 659.3951416015625, "completions/min_length": 227.0, "completions/min_terminated_length": 227.0, "epoch": 0.4744713769984528, "grad_norm": 0.14329631626605988, "kl": 0.01837158203125, "learning_rate": 1e-06, "loss": 0.0176, "num_tokens": 60923858.0, "reward": 1.2178572416305542, "reward_std": 0.18229037523269653, "rewards/code_format_reward/mean": 0.9955357313156128, "rewards/code_format_reward/std": 0.06674052774906158, "rewards/curriculum_aware_reward_fn/mean": 0.22232143580913544, "rewards/curriculum_aware_reward_fn/std": 0.37145575881004333, "step": 115 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1929.0, "completions/max_terminated_length": 1929.0, "completions/mean_length": 657.8192138671875, "completions/mean_terminated_length": 657.8192138671875, "completions/min_length": 189.0, "completions/min_terminated_length": 189.0, "epoch": 0.4785972150593089, "grad_norm": 0.1417432725429535, "kl": 0.020904541015625, "learning_rate": 1e-06, "loss": 0.0069, "num_tokens": 61484015.0, "reward": 1.2268974781036377, "reward_std": 0.17368465662002563, "rewards/code_format_reward/mean": 0.9910714030265808, "rewards/code_format_reward/std": 0.09417349100112915, "rewards/curriculum_aware_reward_fn/mean": 0.23582588136196136, "rewards/curriculum_aware_reward_fn/std": 0.34121277928352356, "step": 116 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 1685.0, "completions/mean_length": 661.5379638671875, "completions/mean_terminated_length": 653.8546142578125, "completions/min_length": 230.0, "completions/min_terminated_length": 230.0, "epoch": 0.48272305312016506, "grad_norm": 0.12646371126174927, "kl": 0.0243072509765625, "learning_rate": 1e-06, "loss": 0.0203, "num_tokens": 62059334.0, "reward": 1.1918528079986572, "reward_std": 0.165035218000412, "rewards/code_format_reward/mean": 0.9910714030265808, "rewards/code_format_reward/std": 0.09417349100112915, "rewards/curriculum_aware_reward_fn/mean": 0.20078125596046448, "rewards/curriculum_aware_reward_fn/std": 0.34070342779159546, "step": 117 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1643.0, "completions/max_terminated_length": 1643.0, "completions/mean_length": 653.8795166015625, "completions/mean_terminated_length": 653.8795166015625, "completions/min_length": 224.0, "completions/min_terminated_length": 224.0, "epoch": 0.48684889118102115, "grad_norm": 0.12753191590309143, "kl": 0.019287109375, "learning_rate": 1e-06, "loss": 0.0139, "num_tokens": 62623494.0, "reward": 1.1503348350524902, "reward_std": 0.14808601140975952, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843171834946, "rewards/curriculum_aware_reward_fn/mean": 0.15703126788139343, "rewards/curriculum_aware_reward_fn/std": 0.28548145294189453, "step": 118 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1760.0, "completions/max_terminated_length": 1760.0, "completions/mean_length": 681.0357666015625, "completions/mean_terminated_length": 681.0357666015625, "completions/min_length": 243.0, "completions/min_terminated_length": 243.0, "epoch": 0.49097472924187724, "grad_norm": 0.12283871322870255, "kl": 0.0186309814453125, "learning_rate": 1e-06, "loss": 0.0044, "num_tokens": 63201052.0, "reward": 1.1948661804199219, "reward_std": 0.14294980466365814, "rewards/code_format_reward/mean": 0.9955357313156128, "rewards/code_format_reward/std": 0.06674052774906158, "rewards/curriculum_aware_reward_fn/mean": 0.19933035969734192, "rewards/curriculum_aware_reward_fn/std": 0.4071090817451477, "step": 119 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 2286.0, "completions/mean_length": 666.0714721679688, "completions/mean_terminated_length": 658.398193359375, "completions/min_length": 275.0, "completions/min_terminated_length": 275.0, "epoch": 0.4951005673027334, "grad_norm": 0.14923040568828583, "kl": 0.0207672119140625, "learning_rate": 1e-06, "loss": 0.0211, "num_tokens": 63762043.0, "reward": 1.200111746788025, "reward_std": 0.19496068358421326, "rewards/code_format_reward/mean": 0.9910714030265808, "rewards/code_format_reward/std": 0.09417349100112915, "rewards/curriculum_aware_reward_fn/mean": 0.20904017984867096, "rewards/curriculum_aware_reward_fn/std": 0.3164607286453247, "step": 120 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 2715.0, "completions/mean_length": 694.357177734375, "completions/mean_terminated_length": 686.7471923828125, "completions/min_length": 271.0, "completions/min_terminated_length": 271.0, "epoch": 0.4992264053635895, "grad_norm": 0.13748040795326233, "kl": 0.0281219482421875, "learning_rate": 1e-06, "loss": 0.0191, "num_tokens": 64340263.0, "reward": 1.223884105682373, "reward_std": 0.19004115462303162, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843171834946, "rewards/curriculum_aware_reward_fn/mean": 0.23058035969734192, "rewards/curriculum_aware_reward_fn/std": 0.3455730080604553, "step": 121 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2336.0, "completions/max_terminated_length": 2336.0, "completions/mean_length": 642.421875, "completions/mean_terminated_length": 642.421875, "completions/min_length": 221.0, "completions/min_terminated_length": 221.0, "epoch": 0.5033522434244456, "grad_norm": 0.15578803420066833, "kl": 0.0207061767578125, "learning_rate": 1e-06, "loss": 0.0114, "num_tokens": 64893963.0, "reward": 1.2465404272079468, "reward_std": 0.2177685797214508, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843171834946, "rewards/curriculum_aware_reward_fn/mean": 0.2532365918159485, "rewards/curriculum_aware_reward_fn/std": 0.35354647040367126, "step": 122 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1666.0, "completions/max_terminated_length": 1666.0, "completions/mean_length": 678.1160888671875, "completions/mean_terminated_length": 678.1160888671875, "completions/min_length": 304.0, "completions/min_terminated_length": 304.0, "epoch": 0.5074780814853017, "grad_norm": 0.12444733083248138, "kl": 0.021026611328125, "learning_rate": 1e-06, "loss": 0.0015, "num_tokens": 65479446.0, "reward": 1.1698660850524902, "reward_std": 0.14799921214580536, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843171834946, "rewards/curriculum_aware_reward_fn/mean": 0.17656251788139343, "rewards/curriculum_aware_reward_fn/std": 0.31459254026412964, "step": 123 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1505.0, "completions/max_terminated_length": 1505.0, "completions/mean_length": 659.5379638671875, "completions/mean_terminated_length": 659.5379638671875, "completions/min_length": 195.0, "completions/min_terminated_length": 195.0, "epoch": 0.5116039195461578, "grad_norm": 0.12079131603240967, "kl": 0.0192108154296875, "learning_rate": 1e-06, "loss": 0.0057, "num_tokens": 66030986.0, "reward": 1.1678571701049805, "reward_std": 0.12463776767253876, "rewards/code_format_reward/mean": 0.9977678656578064, "rewards/code_format_reward/std": 0.047245558351278305, "rewards/curriculum_aware_reward_fn/mean": 0.17008928954601288, "rewards/curriculum_aware_reward_fn/std": 0.3068173825740814, "step": 124 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2301.0, "completions/max_terminated_length": 2301.0, "completions/mean_length": 685.2455444335938, "completions/mean_terminated_length": 685.2455444335938, "completions/min_length": 212.0, "completions/min_terminated_length": 212.0, "epoch": 0.5157297576070139, "grad_norm": 0.13699984550476074, "kl": 0.0224761962890625, "learning_rate": 1e-06, "loss": 0.0124, "num_tokens": 66600462.0, "reward": 1.1631696224212646, "reward_std": 0.164546936750412, "rewards/code_format_reward/mean": 0.9866071343421936, "rewards/code_format_reward/std": 0.11507843434810638, "rewards/curriculum_aware_reward_fn/mean": 0.17656250298023224, "rewards/curriculum_aware_reward_fn/std": 0.30818045139312744, "step": 125 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2236.0, "completions/max_terminated_length": 2236.0, "completions/mean_length": 631.8861694335938, "completions/mean_terminated_length": 631.8861694335938, "completions/min_length": 186.0, "completions/min_terminated_length": 186.0, "epoch": 0.51985559566787, "grad_norm": 0.13626828789710999, "kl": 0.021026611328125, "learning_rate": 1e-06, "loss": 0.0092, "num_tokens": 67142960.0, "reward": 1.1934152841567993, "reward_std": 0.139185830950737, "rewards/code_format_reward/mean": 0.9977678656578064, "rewards/code_format_reward/std": 0.047245558351278305, "rewards/curriculum_aware_reward_fn/mean": 0.19564732909202576, "rewards/curriculum_aware_reward_fn/std": 0.32974809408187866, "step": 126 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1879.0, "completions/max_terminated_length": 1879.0, "completions/mean_length": 669.2053833007812, "completions/mean_terminated_length": 669.2053833007812, "completions/min_length": 239.0, "completions/min_terminated_length": 239.0, "epoch": 0.5239814337287262, "grad_norm": 0.1478445678949356, "kl": 0.019927978515625, "learning_rate": 1e-06, "loss": 0.0028, "num_tokens": 67716170.0, "reward": 1.225334882736206, "reward_std": 0.187877357006073, "rewards/code_format_reward/mean": 0.9977678656578064, "rewards/code_format_reward/std": 0.047245558351278305, "rewards/curriculum_aware_reward_fn/mean": 0.22756695747375488, "rewards/curriculum_aware_reward_fn/std": 0.3380330204963684, "step": 127 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2743.0, "completions/max_terminated_length": 2743.0, "completions/mean_length": 682.1942138671875, "completions/mean_terminated_length": 682.1942138671875, "completions/min_length": 260.0, "completions/min_terminated_length": 260.0, "epoch": 0.5281072717895823, "grad_norm": 0.12930497527122498, "kl": 0.0203094482421875, "learning_rate": 1e-06, "loss": 0.0058, "num_tokens": 68288444.0, "reward": 1.1726562976837158, "reward_std": 0.14786198735237122, "rewards/code_format_reward/mean": 0.9866071343421936, "rewards/code_format_reward/std": 0.11507844179868698, "rewards/curriculum_aware_reward_fn/mean": 0.18604911863803864, "rewards/curriculum_aware_reward_fn/std": 0.31647637486457825, "step": 128 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1721.0, "completions/max_terminated_length": 1721.0, "completions/mean_length": 703.4642944335938, "completions/mean_terminated_length": 703.4642944335938, "completions/min_length": 361.0, "completions/min_terminated_length": 361.0, "epoch": 0.5322331098504384, "grad_norm": 0.1500132977962494, "kl": 0.020538330078125, "learning_rate": 1e-06, "loss": -0.0023, "num_tokens": 68901565.0, "reward": 1.2012277841567993, "reward_std": 0.2054479569196701, "rewards/code_format_reward/mean": 0.9977678656578064, "rewards/code_format_reward/std": 0.047245558351278305, "rewards/curriculum_aware_reward_fn/mean": 0.20345982909202576, "rewards/curriculum_aware_reward_fn/std": 0.3204866945743561, "step": 129 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1989.0, "completions/max_terminated_length": 1989.0, "completions/mean_length": 679.2299194335938, "completions/mean_terminated_length": 679.2299194335938, "completions/min_length": 229.0, "completions/min_terminated_length": 229.0, "epoch": 0.5363589479112945, "grad_norm": 0.15506230294704437, "kl": 0.02056884765625, "learning_rate": 1e-06, "loss": 0.028, "num_tokens": 69472504.0, "reward": 1.2210938930511475, "reward_std": 0.2002020925283432, "rewards/code_format_reward/mean": 0.9888392686843872, "rewards/code_format_reward/std": 0.10517053306102753, "rewards/curriculum_aware_reward_fn/mean": 0.23225447535514832, "rewards/curriculum_aware_reward_fn/std": 0.3653670847415924, "step": 130 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 1793.0, "completions/mean_length": 693.5178833007812, "completions/mean_terminated_length": 685.9060668945312, "completions/min_length": 264.0, "completions/min_terminated_length": 264.0, "epoch": 0.5404847859721505, "grad_norm": 0.14253978431224823, "kl": 0.0223388671875, "learning_rate": 1e-06, "loss": 0.0173, "num_tokens": 70061205.0, "reward": 1.183370590209961, "reward_std": 0.18808096647262573, "rewards/code_format_reward/mean": 0.9821428656578064, "rewards/code_format_reward/std": 0.13258016109466553, "rewards/curriculum_aware_reward_fn/mean": 0.20122769474983215, "rewards/curriculum_aware_reward_fn/std": 0.32476744055747986, "step": 131 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 1658.0, "completions/mean_length": 711.3795166015625, "completions/mean_terminated_length": 703.8076171875, "completions/min_length": 202.0, "completions/min_terminated_length": 202.0, "epoch": 0.5446106240330068, "grad_norm": 0.11306580901145935, "kl": 0.01953125, "learning_rate": 1e-06, "loss": 0.0036, "num_tokens": 70662430.0, "reward": 1.141852855682373, "reward_std": 0.11286230385303497, "rewards/code_format_reward/mean": 0.9977678656578064, "rewards/code_format_reward/std": 0.047245558351278305, "rewards/curriculum_aware_reward_fn/mean": 0.14408482611179352, "rewards/curriculum_aware_reward_fn/std": 0.2872644364833832, "step": 132 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1500.0, "completions/max_terminated_length": 1500.0, "completions/mean_length": 677.6116333007812, "completions/mean_terminated_length": 677.6116333007812, "completions/min_length": 196.0, "completions/min_terminated_length": 196.0, "epoch": 0.5487364620938628, "grad_norm": 0.137791708111763, "kl": 0.0191192626953125, "learning_rate": 1e-06, "loss": 0.0102, "num_tokens": 71244467.0, "reward": 1.1542412042617798, "reward_std": 0.15288133919239044, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843916893005, "rewards/curriculum_aware_reward_fn/mean": 0.16093751788139343, "rewards/curriculum_aware_reward_fn/std": 0.3050521910190582, "step": 133 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2347.0, "completions/max_terminated_length": 2347.0, "completions/mean_length": 720.3817138671875, "completions/mean_terminated_length": 720.3817138671875, "completions/min_length": 253.0, "completions/min_terminated_length": 253.0, "epoch": 0.5528623001547189, "grad_norm": 0.1263774335384369, "kl": 0.0180511474609375, "learning_rate": 1e-06, "loss": 0.0104, "num_tokens": 71837706.0, "reward": 1.1943081617355347, "reward_std": 0.16193082928657532, "rewards/code_format_reward/mean": 0.9977678656578064, "rewards/code_format_reward/std": 0.047245558351278305, "rewards/curriculum_aware_reward_fn/mean": 0.19654019176959991, "rewards/curriculum_aware_reward_fn/std": 0.31083038449287415, "step": 134 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 3160.0, "completions/max_terminated_length": 3160.0, "completions/mean_length": 717.8170166015625, "completions/mean_terminated_length": 717.8170166015625, "completions/min_length": 255.0, "completions/min_terminated_length": 255.0, "epoch": 0.556988138215575, "grad_norm": 0.13417816162109375, "kl": 0.01806640625, "learning_rate": 1e-06, "loss": -0.0025, "num_tokens": 72442956.0, "reward": 1.2074778079986572, "reward_std": 0.1575753092765808, "rewards/code_format_reward/mean": 1.0, "rewards/code_format_reward/std": 0.0, "rewards/curriculum_aware_reward_fn/mean": 0.20747768878936768, "rewards/curriculum_aware_reward_fn/std": 0.3245628774166107, "step": 135 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1862.0, "completions/max_terminated_length": 1862.0, "completions/mean_length": 689.685302734375, "completions/mean_terminated_length": 689.685302734375, "completions/min_length": 304.0, "completions/min_terminated_length": 304.0, "epoch": 0.5611139762764311, "grad_norm": 0.14430135488510132, "kl": 0.0204010009765625, "learning_rate": 1e-06, "loss": 0.0019, "num_tokens": 73013770.0, "reward": 1.203125, "reward_std": 0.1807006150484085, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843171834946, "rewards/curriculum_aware_reward_fn/mean": 0.2098214477300644, "rewards/curriculum_aware_reward_fn/std": 0.3212871849536896, "step": 136 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1962.0, "completions/max_terminated_length": 1962.0, "completions/mean_length": 722.122802734375, "completions/mean_terminated_length": 722.122802734375, "completions/min_length": 226.0, "completions/min_terminated_length": 226.0, "epoch": 0.5652398143372873, "grad_norm": 0.11680962890386581, "kl": 0.0230712890625, "learning_rate": 1e-06, "loss": -0.0045, "num_tokens": 73602791.0, "reward": 1.1407365798950195, "reward_std": 0.15920059382915497, "rewards/code_format_reward/mean": 0.9910714030265808, "rewards/code_format_reward/std": 0.09417349100112915, "rewards/curriculum_aware_reward_fn/mean": 0.14966519176959991, "rewards/curriculum_aware_reward_fn/std": 0.3030141294002533, "step": 137 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 2287.0, "completions/mean_length": 739.935302734375, "completions/mean_terminated_length": 732.4273071289062, "completions/min_length": 298.0, "completions/min_terminated_length": 298.0, "epoch": 0.5693656523981434, "grad_norm": 0.13728173077106476, "kl": 0.019561767578125, "learning_rate": 1e-06, "loss": -0.0027, "num_tokens": 74210215.0, "reward": 1.2183036804199219, "reward_std": 0.17961503565311432, "rewards/code_format_reward/mean": 0.9955357313156128, "rewards/code_format_reward/std": 0.06674052774906158, "rewards/curriculum_aware_reward_fn/mean": 0.22276785969734192, "rewards/curriculum_aware_reward_fn/std": 0.3419995605945587, "step": 138 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 3354.0, "completions/mean_length": 724.7991333007812, "completions/mean_terminated_length": 717.2572631835938, "completions/min_length": 231.0, "completions/min_terminated_length": 231.0, "epoch": 0.5734914904589995, "grad_norm": 0.1271560788154602, "kl": 0.0178070068359375, "learning_rate": 1e-06, "loss": 0.0136, "num_tokens": 74785686.0, "reward": 1.191183090209961, "reward_std": 0.18125225603580475, "rewards/code_format_reward/mean": 0.9888392686843872, "rewards/code_format_reward/std": 0.10517053306102753, "rewards/curriculum_aware_reward_fn/mean": 0.20234374701976776, "rewards/curriculum_aware_reward_fn/std": 0.33361297845840454, "step": 139 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1970.0, "completions/max_terminated_length": 1970.0, "completions/mean_length": 745.3928833007812, "completions/mean_terminated_length": 745.3928833007812, "completions/min_length": 325.0, "completions/min_terminated_length": 325.0, "epoch": 0.5776173285198556, "grad_norm": 0.10332006961107254, "kl": 0.0187835693359375, "learning_rate": 1e-06, "loss": 0.0014, "num_tokens": 75394974.0, "reward": 1.145759105682373, "reward_std": 0.12595143914222717, "rewards/code_format_reward/mean": 0.9955357313156128, "rewards/code_format_reward/std": 0.06674052774906158, "rewards/curriculum_aware_reward_fn/mean": 0.15022322535514832, "rewards/curriculum_aware_reward_fn/std": 0.31267043948173523, "step": 140 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2256.0, "completions/max_terminated_length": 2256.0, "completions/mean_length": 738.8326416015625, "completions/mean_terminated_length": 738.8326416015625, "completions/min_length": 313.0, "completions/min_terminated_length": 313.0, "epoch": 0.5817431665807117, "grad_norm": 0.13378408551216125, "kl": 0.0215606689453125, "learning_rate": 1e-06, "loss": -0.001, "num_tokens": 75996689.0, "reward": 1.248549222946167, "reward_std": 0.18088890612125397, "rewards/code_format_reward/mean": 0.9977678656578064, "rewards/code_format_reward/std": 0.047245558351278305, "rewards/curriculum_aware_reward_fn/mean": 0.25078126788139343, "rewards/curriculum_aware_reward_fn/std": 0.35256245732307434, "step": 141 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 1841.0, "completions/mean_length": 810.1808471679688, "completions/mean_terminated_length": 802.8299560546875, "completions/min_length": 239.0, "completions/min_terminated_length": 239.0, "epoch": 0.5858690046415678, "grad_norm": 0.1341581493616104, "kl": 0.018035888671875, "learning_rate": 1e-06, "loss": 0.0081, "num_tokens": 76649408.0, "reward": 1.208147406578064, "reward_std": 0.19399170577526093, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843916893005, "rewards/curriculum_aware_reward_fn/mean": 0.2148437649011612, "rewards/curriculum_aware_reward_fn/std": 0.3443984091281891, "step": 142 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1818.0, "completions/max_terminated_length": 1818.0, "completions/mean_length": 781.5803833007812, "completions/mean_terminated_length": 781.5803833007812, "completions/min_length": 314.0, "completions/min_terminated_length": 314.0, "epoch": 0.589994842702424, "grad_norm": 0.132659450173378, "kl": 0.01953125, "learning_rate": 1e-06, "loss": -0.0034, "num_tokens": 77281019.0, "reward": 1.1776787042617798, "reward_std": 0.17381568253040314, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843916893005, "rewards/curriculum_aware_reward_fn/mean": 0.18437501788139343, "rewards/curriculum_aware_reward_fn/std": 0.30239763855934143, "step": 143 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 3096.0, "completions/max_terminated_length": 3096.0, "completions/mean_length": 785.8348388671875, "completions/mean_terminated_length": 785.8348388671875, "completions/min_length": 240.0, "completions/min_terminated_length": 240.0, "epoch": 0.5941206807632801, "grad_norm": 0.11722385883331299, "kl": 0.0221405029296875, "learning_rate": 1e-06, "loss": -0.0007, "num_tokens": 77892136.0, "reward": 1.1713169813156128, "reward_std": 0.18527261912822723, "rewards/code_format_reward/mean": 0.9888392686843872, "rewards/code_format_reward/std": 0.10517053306102753, "rewards/curriculum_aware_reward_fn/mean": 0.1824776977300644, "rewards/curriculum_aware_reward_fn/std": 0.3004841208457947, "step": 144 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2478.0, "completions/max_terminated_length": 2478.0, "completions/mean_length": 837.3795166015625, "completions/mean_terminated_length": 837.3795166015625, "completions/min_length": 350.0, "completions/min_terminated_length": 350.0, "epoch": 0.5982465188241362, "grad_norm": 0.12240413576364517, "kl": 0.01800537109375, "learning_rate": 1e-06, "loss": 0.0177, "num_tokens": 78536621.0, "reward": 1.1825892925262451, "reward_std": 0.17582431435585022, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843171834946, "rewards/curriculum_aware_reward_fn/mean": 0.18928572535514832, "rewards/curriculum_aware_reward_fn/std": 0.3062494993209839, "step": 145 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2151.0, "completions/max_terminated_length": 2151.0, "completions/mean_length": 795.5022583007812, "completions/mean_terminated_length": 795.5022583007812, "completions/min_length": 236.0, "completions/min_terminated_length": 236.0, "epoch": 0.6023723568849922, "grad_norm": 0.136099711060524, "kl": 0.0181121826171875, "learning_rate": 1e-06, "loss": 0.0086, "num_tokens": 79148688.0, "reward": 1.2521207332611084, "reward_std": 0.2308141589164734, "rewards/code_format_reward/mean": 0.9866071343421936, "rewards/code_format_reward/std": 0.11507843434810638, "rewards/curriculum_aware_reward_fn/mean": 0.2655133903026581, "rewards/curriculum_aware_reward_fn/std": 0.38686394691467285, "step": 146 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 2230.0, "completions/mean_length": 918.5826416015625, "completions/mean_terminated_length": 911.4743041992188, "completions/min_length": 344.0, "completions/min_terminated_length": 344.0, "epoch": 0.6064981949458483, "grad_norm": 0.12820293009281158, "kl": 0.019775390625, "learning_rate": 1e-06, "loss": 0.0175, "num_tokens": 79826017.0, "reward": 1.2609375715255737, "reward_std": 0.26225149631500244, "rewards/code_format_reward/mean": 0.9866071343421936, "rewards/code_format_reward/std": 0.11507843434810638, "rewards/curriculum_aware_reward_fn/mean": 0.27433037757873535, "rewards/curriculum_aware_reward_fn/std": 0.3973761796951294, "step": 147 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 3542.0, "completions/max_terminated_length": 3542.0, "completions/mean_length": 868.1897583007812, "completions/mean_terminated_length": 868.1897583007812, "completions/min_length": 290.0, "completions/min_terminated_length": 290.0, "epoch": 0.6106240330067045, "grad_norm": 0.1212223693728447, "kl": 0.017730712890625, "learning_rate": 1e-06, "loss": 0.0039, "num_tokens": 80486566.0, "reward": 1.1830357313156128, "reward_std": 0.166142538189888, "rewards/code_format_reward/mean": 0.9955357313156128, "rewards/code_format_reward/std": 0.06674052774906158, "rewards/curriculum_aware_reward_fn/mean": 0.1875, "rewards/curriculum_aware_reward_fn/std": 0.3334619402885437, "step": 148 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 3657.0, "completions/max_terminated_length": 3657.0, "completions/mean_length": 866.654052734375, "completions/mean_terminated_length": 866.654052734375, "completions/min_length": 299.0, "completions/min_terminated_length": 299.0, "epoch": 0.6147498710675606, "grad_norm": 0.11715445667505264, "kl": 0.021759033203125, "learning_rate": 1e-06, "loss": 0.007, "num_tokens": 81163262.0, "reward": 1.2314733266830444, "reward_std": 0.16743095219135284, "rewards/code_format_reward/mean": 0.9888392686843872, "rewards/code_format_reward/std": 0.10517053306102753, "rewards/curriculum_aware_reward_fn/mean": 0.24263392388820648, "rewards/curriculum_aware_reward_fn/std": 0.3736143410205841, "step": 149 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2317.0, "completions/max_terminated_length": 2317.0, "completions/mean_length": 940.49560546875, "completions/mean_terminated_length": 940.49560546875, "completions/min_length": 333.0, "completions/min_terminated_length": 333.0, "epoch": 0.6188757091284167, "grad_norm": 0.11408567428588867, "kl": 0.018310546875, "learning_rate": 1e-06, "loss": 0.0101, "num_tokens": 81870295.0, "reward": 1.1629464626312256, "reward_std": 0.19442030787467957, "rewards/code_format_reward/mean": 0.9888392686843872, "rewards/code_format_reward/std": 0.10517053306102753, "rewards/curriculum_aware_reward_fn/mean": 0.1741071492433548, "rewards/curriculum_aware_reward_fn/std": 0.31434616446495056, "step": 150 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 1966.0, "completions/mean_length": 859.3058471679688, "completions/mean_terminated_length": 852.0648803710938, "completions/min_length": 248.0, "completions/min_terminated_length": 248.0, "epoch": 0.6230015471892728, "grad_norm": 0.11013923585414886, "kl": 0.019134521484375, "learning_rate": 1e-06, "loss": 0.0234, "num_tokens": 82518352.0, "reward": 1.1734375953674316, "reward_std": 0.15971694886684418, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843171834946, "rewards/curriculum_aware_reward_fn/mean": 0.18013392388820648, "rewards/curriculum_aware_reward_fn/std": 0.3331034481525421, "step": 151 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 2645.0, "completions/mean_length": 884.8750610351562, "completions/mean_terminated_length": 877.6912841796875, "completions/min_length": 283.0, "completions/min_terminated_length": 283.0, "epoch": 0.6271273852501289, "grad_norm": 0.11793754994869232, "kl": 0.01857757568359375, "learning_rate": 1e-06, "loss": 0.014, "num_tokens": 83171718.0, "reward": 1.2364956140518188, "reward_std": 0.2031078040599823, "rewards/code_format_reward/mean": 0.9888392686843872, "rewards/code_format_reward/std": 0.10517053306102753, "rewards/curriculum_aware_reward_fn/mean": 0.24765624105930328, "rewards/curriculum_aware_reward_fn/std": 0.381111741065979, "step": 152 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2032.0, "completions/max_terminated_length": 2032.0, "completions/mean_length": 906.779052734375, "completions/mean_terminated_length": 906.779052734375, "completions/min_length": 356.0, "completions/min_terminated_length": 356.0, "epoch": 0.631253223310985, "grad_norm": 0.12254896014928818, "kl": 0.02001953125, "learning_rate": 1e-06, "loss": 0.0226, "num_tokens": 83852292.0, "reward": 1.2754465341567993, "reward_std": 0.22218114137649536, "rewards/code_format_reward/mean": 0.9888392686843872, "rewards/code_format_reward/std": 0.10517053306102753, "rewards/curriculum_aware_reward_fn/mean": 0.28660711646080017, "rewards/curriculum_aware_reward_fn/std": 0.39385560154914856, "step": 153 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2647.0, "completions/max_terminated_length": 2647.0, "completions/mean_length": 948.1004638671875, "completions/mean_terminated_length": 948.1004638671875, "completions/min_length": 324.0, "completions/min_terminated_length": 324.0, "epoch": 0.6353790613718412, "grad_norm": 0.11475897580385208, "kl": 0.017730712890625, "learning_rate": 1e-06, "loss": -0.0026, "num_tokens": 84546982.0, "reward": 1.190178632736206, "reward_std": 0.17127592861652374, "rewards/code_format_reward/mean": 0.9955357313156128, "rewards/code_format_reward/std": 0.06674052774906158, "rewards/curriculum_aware_reward_fn/mean": 0.19464285671710968, "rewards/curriculum_aware_reward_fn/std": 0.33930492401123047, "step": 154 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2183.0, "completions/max_terminated_length": 2183.0, "completions/mean_length": 968.810302734375, "completions/mean_terminated_length": 968.810302734375, "completions/min_length": 329.0, "completions/min_terminated_length": 329.0, "epoch": 0.6395048994326973, "grad_norm": 0.09638718515634537, "kl": 0.017425537109375, "learning_rate": 1e-06, "loss": 0.0117, "num_tokens": 85263592.0, "reward": 1.1160714626312256, "reward_std": 0.12120134383440018, "rewards/code_format_reward/mean": 0.9888392686843872, "rewards/code_format_reward/std": 0.10517053306102753, "rewards/curriculum_aware_reward_fn/mean": 0.1272321492433548, "rewards/curriculum_aware_reward_fn/std": 0.2862734794616699, "step": 155 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2999.0, "completions/max_terminated_length": 2999.0, "completions/mean_length": 908.7388916015625, "completions/mean_terminated_length": 908.7388916015625, "completions/min_length": 304.0, "completions/min_terminated_length": 304.0, "epoch": 0.6436307374935534, "grad_norm": 0.1060652881860733, "kl": 0.019561767578125, "learning_rate": 1e-06, "loss": 0.006, "num_tokens": 85946143.0, "reward": 1.1916296482086182, "reward_std": 0.17853893339633942, "rewards/code_format_reward/mean": 0.9866071343421936, "rewards/code_format_reward/std": 0.11507843434810638, "rewards/curriculum_aware_reward_fn/mean": 0.20502232015132904, "rewards/curriculum_aware_reward_fn/std": 0.36013975739479065, "step": 156 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 1950.0, "completions/mean_length": 937.450927734375, "completions/mean_terminated_length": 930.384765625, "completions/min_length": 290.0, "completions/min_terminated_length": 290.0, "epoch": 0.6477565755544095, "grad_norm": 0.10560503602027893, "kl": 0.017791748046875, "learning_rate": 1e-06, "loss": 0.0086, "num_tokens": 86643734.0, "reward": 1.1947544813156128, "reward_std": 0.16470228135585785, "rewards/code_format_reward/mean": 0.984375, "rewards/code_format_reward/std": 0.12415824085474014, "rewards/curriculum_aware_reward_fn/mean": 0.2103794664144516, "rewards/curriculum_aware_reward_fn/std": 0.4054417908191681, "step": 157 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2555.0, "completions/max_terminated_length": 2555.0, "completions/mean_length": 951.1160888671875, "completions/mean_terminated_length": 951.1160888671875, "completions/min_length": 372.0, "completions/min_terminated_length": 372.0, "epoch": 0.6518824136152656, "grad_norm": 0.10858561098575592, "kl": 0.018768310546875, "learning_rate": 1e-06, "loss": -0.002, "num_tokens": 87314057.0, "reward": 1.1876118183135986, "reward_std": 0.16939528286457062, "rewards/code_format_reward/mean": 0.9866071343421936, "rewards/code_format_reward/std": 0.11507843434810638, "rewards/curriculum_aware_reward_fn/mean": 0.20100447535514832, "rewards/curriculum_aware_reward_fn/std": 0.3483969271183014, "step": 158 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 3312.0, "completions/mean_length": 960.6428833007812, "completions/mean_terminated_length": 953.628662109375, "completions/min_length": 267.0, "completions/min_terminated_length": 267.0, "epoch": 0.6560082516761218, "grad_norm": 0.10794656723737717, "kl": 0.0206146240234375, "learning_rate": 1e-06, "loss": 0.0063, "num_tokens": 88024838.0, "reward": 1.1988840103149414, "reward_std": 0.1717086285352707, "rewards/code_format_reward/mean": 0.9866071343421936, "rewards/code_format_reward/std": 0.11507844179868698, "rewards/curriculum_aware_reward_fn/mean": 0.21227678656578064, "rewards/curriculum_aware_reward_fn/std": 0.3667148947715759, "step": 159 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 3339.0, "completions/max_terminated_length": 3339.0, "completions/mean_length": 998.7701416015625, "completions/mean_terminated_length": 998.7701416015625, "completions/min_length": 249.0, "completions/min_terminated_length": 249.0, "epoch": 0.6601340897369778, "grad_norm": 0.10519266873598099, "kl": 0.01983642578125, "learning_rate": 1e-06, "loss": 0.0052, "num_tokens": 88772153.0, "reward": 1.1863839626312256, "reward_std": 0.1970013678073883, "rewards/code_format_reward/mean": 0.9799107313156128, "rewards/code_format_reward/std": 0.14046260714530945, "rewards/curriculum_aware_reward_fn/mean": 0.2064732164144516, "rewards/curriculum_aware_reward_fn/std": 0.35936254262924194, "step": 160 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2166.0, "completions/max_terminated_length": 2166.0, "completions/mean_length": 947.63623046875, "completions/mean_terminated_length": 947.63623046875, "completions/min_length": 451.0, "completions/min_terminated_length": 451.0, "epoch": 0.6642599277978339, "grad_norm": 0.10040335357189178, "kl": 0.0175018310546875, "learning_rate": 1e-06, "loss": -0.0027, "num_tokens": 89469330.0, "reward": 1.1856027841567993, "reward_std": 0.16616688668727875, "rewards/code_format_reward/mean": 0.984375, "rewards/code_format_reward/std": 0.12415824085474014, "rewards/curriculum_aware_reward_fn/mean": 0.20122767984867096, "rewards/curriculum_aware_reward_fn/std": 0.3571789562702179, "step": 161 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2180.0, "completions/max_terminated_length": 2180.0, "completions/mean_length": 924.0223388671875, "completions/mean_terminated_length": 924.0223388671875, "completions/min_length": 279.0, "completions/min_terminated_length": 279.0, "epoch": 0.66838576585869, "grad_norm": 0.11762242764234543, "kl": 0.0186004638671875, "learning_rate": 1e-06, "loss": 0.0044, "num_tokens": 90152183.0, "reward": 1.240625023841858, "reward_std": 0.20828305184841156, "rewards/code_format_reward/mean": 0.9866071343421936, "rewards/code_format_reward/std": 0.11507843434810638, "rewards/curriculum_aware_reward_fn/mean": 0.2540178596973419, "rewards/curriculum_aware_reward_fn/std": 0.3628675639629364, "step": 162 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2780.0, "completions/max_terminated_length": 2780.0, "completions/mean_length": 946.8058471679688, "completions/mean_terminated_length": 946.8058471679688, "completions/min_length": 422.0, "completions/min_terminated_length": 422.0, "epoch": 0.6725116039195461, "grad_norm": 0.10841399431228638, "kl": 0.01715087890625, "learning_rate": 1e-06, "loss": 0.0096, "num_tokens": 90843192.0, "reward": 1.2158483266830444, "reward_std": 0.2003968209028244, "rewards/code_format_reward/mean": 0.984375, "rewards/code_format_reward/std": 0.12415824085474014, "rewards/curriculum_aware_reward_fn/mean": 0.23147322237491608, "rewards/curriculum_aware_reward_fn/std": 0.37479665875434875, "step": 163 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 3465.0, "completions/mean_length": 917.43310546875, "completions/mean_terminated_length": 910.3221435546875, "completions/min_length": 393.0, "completions/min_terminated_length": 393.0, "epoch": 0.6766374419804023, "grad_norm": 0.10595975816249847, "kl": 0.0200653076171875, "learning_rate": 1e-06, "loss": 0.0179, "num_tokens": 91524801.0, "reward": 1.1857142448425293, "reward_std": 0.19158855080604553, "rewards/code_format_reward/mean": 0.984375, "rewards/code_format_reward/std": 0.12415824085474014, "rewards/curriculum_aware_reward_fn/mean": 0.20133927464485168, "rewards/curriculum_aware_reward_fn/std": 0.34559935331344604, "step": 164 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 3099.0, "completions/max_terminated_length": 3099.0, "completions/mean_length": 949.7567138671875, "completions/mean_terminated_length": 949.7567138671875, "completions/min_length": 398.0, "completions/min_terminated_length": 398.0, "epoch": 0.6807632800412584, "grad_norm": 0.12277974933385849, "kl": 0.0175018310546875, "learning_rate": 1e-06, "loss": -0.0024, "num_tokens": 92230005.0, "reward": 1.2635046243667603, "reward_std": 0.2197643518447876, "rewards/code_format_reward/mean": 0.9866071343421936, "rewards/code_format_reward/std": 0.11507843434810638, "rewards/curriculum_aware_reward_fn/mean": 0.2768973410129547, "rewards/curriculum_aware_reward_fn/std": 0.3969356417655945, "step": 165 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1808.0, "completions/max_terminated_length": 1808.0, "completions/mean_length": 939.0000610351562, "completions/mean_terminated_length": 939.0000610351562, "completions/min_length": 304.0, "completions/min_terminated_length": 304.0, "epoch": 0.6848891181021145, "grad_norm": 0.10809571295976639, "kl": 0.02056884765625, "learning_rate": 1e-06, "loss": 0.0025, "num_tokens": 92935892.0, "reward": 1.1876118183135986, "reward_std": 0.18095649778842926, "rewards/code_format_reward/mean": 0.984375, "rewards/code_format_reward/std": 0.12415824085474014, "rewards/curriculum_aware_reward_fn/mean": 0.20323660969734192, "rewards/curriculum_aware_reward_fn/std": 0.35159531235694885, "step": 166 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2375.0, "completions/max_terminated_length": 2375.0, "completions/mean_length": 891.779052734375, "completions/mean_terminated_length": 891.779052734375, "completions/min_length": 385.0, "completions/min_terminated_length": 385.0, "epoch": 0.6890149561629706, "grad_norm": 0.1229667067527771, "kl": 0.019012451171875, "learning_rate": 1e-06, "loss": 0.0041, "num_tokens": 93601787.0, "reward": 1.2900670766830444, "reward_std": 0.2178320288658142, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843916893005, "rewards/curriculum_aware_reward_fn/mean": 0.2967633903026581, "rewards/curriculum_aware_reward_fn/std": 0.39798933267593384, "step": 167 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2851.0, "completions/max_terminated_length": 2851.0, "completions/mean_length": 1009.2813110351562, "completions/mean_terminated_length": 1009.2813110351562, "completions/min_length": 401.0, "completions/min_terminated_length": 401.0, "epoch": 0.6931407942238267, "grad_norm": 0.0931694358587265, "kl": 0.0195159912109375, "learning_rate": 1e-06, "loss": 0.0008, "num_tokens": 94344613.0, "reward": 1.1193081140518188, "reward_std": 0.11687362939119339, "rewards/code_format_reward/mean": 0.9977678656578064, "rewards/code_format_reward/std": 0.047245558351278305, "rewards/curriculum_aware_reward_fn/mean": 0.12154017388820648, "rewards/curriculum_aware_reward_fn/std": 0.27965956926345825, "step": 168 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2592.0, "completions/max_terminated_length": 2592.0, "completions/mean_length": 962.22998046875, "completions/mean_terminated_length": 962.22998046875, "completions/min_length": 383.0, "completions/min_terminated_length": 383.0, "epoch": 0.6972666322846828, "grad_norm": 0.09708420932292938, "kl": 0.019500732421875, "learning_rate": 1e-06, "loss": 0.0101, "num_tokens": 95061145.0, "reward": 1.1991071701049805, "reward_std": 0.14699062705039978, "rewards/code_format_reward/mean": 0.9888392686843872, "rewards/code_format_reward/std": 0.10517053306102753, "rewards/curriculum_aware_reward_fn/mean": 0.21026785671710968, "rewards/curriculum_aware_reward_fn/std": 0.3594140410423279, "step": 169 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2848.0, "completions/max_terminated_length": 2848.0, "completions/mean_length": 944.4710083007812, "completions/mean_terminated_length": 944.4710083007812, "completions/min_length": 386.0, "completions/min_terminated_length": 386.0, "epoch": 0.701392470345539, "grad_norm": 0.10557137429714203, "kl": 0.0189971923828125, "learning_rate": 1e-06, "loss": 0.0099, "num_tokens": 95776462.0, "reward": 1.1698662042617798, "reward_std": 0.11741764843463898, "rewards/code_format_reward/mean": 0.9955357313156128, "rewards/code_format_reward/std": 0.06674052774906158, "rewards/curriculum_aware_reward_fn/mean": 0.17433033883571625, "rewards/curriculum_aware_reward_fn/std": 0.31493332982063293, "step": 170 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2458.0, "completions/max_terminated_length": 2458.0, "completions/mean_length": 955.6160888671875, "completions/mean_terminated_length": 955.6160888671875, "completions/min_length": 303.0, "completions/min_terminated_length": 303.0, "epoch": 0.7055183084063951, "grad_norm": 0.10741835832595825, "kl": 0.01788330078125, "learning_rate": 1e-06, "loss": 0.0052, "num_tokens": 96478397.0, "reward": 1.2027901411056519, "reward_std": 0.1658809781074524, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843916893005, "rewards/curriculum_aware_reward_fn/mean": 0.20948661863803864, "rewards/curriculum_aware_reward_fn/std": 0.3418176770210266, "step": 171 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2378.0, "completions/max_terminated_length": 2378.0, "completions/mean_length": 937.122802734375, "completions/mean_terminated_length": 937.122802734375, "completions/min_length": 363.0, "completions/min_terminated_length": 363.0, "epoch": 0.7096441464672512, "grad_norm": 0.10694112628698349, "kl": 0.0186920166015625, "learning_rate": 1e-06, "loss": -0.0023, "num_tokens": 97158670.0, "reward": 1.2568081617355347, "reward_std": 0.19249658286571503, "rewards/code_format_reward/mean": 0.9821428656578064, "rewards/code_format_reward/std": 0.13258016109466553, "rewards/curriculum_aware_reward_fn/mean": 0.2746651768684387, "rewards/curriculum_aware_reward_fn/std": 0.4220302104949951, "step": 172 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2915.0, "completions/max_terminated_length": 2915.0, "completions/mean_length": 929.96435546875, "completions/mean_terminated_length": 929.96435546875, "completions/min_length": 323.0, "completions/min_terminated_length": 323.0, "epoch": 0.7137699845281072, "grad_norm": 0.0922522023320198, "kl": 0.01833343505859375, "learning_rate": 1e-06, "loss": -0.0006, "num_tokens": 97832215.0, "reward": 1.1475446224212646, "reward_std": 0.12352091073989868, "rewards/code_format_reward/mean": 0.9910714030265808, "rewards/code_format_reward/std": 0.09417349100112915, "rewards/curriculum_aware_reward_fn/mean": 0.15647320449352264, "rewards/curriculum_aware_reward_fn/std": 0.3055644631385803, "step": 173 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 2144.0, "completions/mean_length": 901.3594360351562, "completions/mean_terminated_length": 894.2125244140625, "completions/min_length": 266.0, "completions/min_terminated_length": 266.0, "epoch": 0.7178958225889633, "grad_norm": 0.1305251121520996, "kl": 0.0187530517578125, "learning_rate": 1e-06, "loss": 0.0112, "num_tokens": 98502591.0, "reward": 1.3667412996292114, "reward_std": 0.2752566933631897, "rewards/code_format_reward/mean": 0.9977678656578064, "rewards/code_format_reward/std": 0.047245558351278305, "rewards/curriculum_aware_reward_fn/mean": 0.3689732551574707, "rewards/curriculum_aware_reward_fn/std": 0.4514053165912628, "step": 174 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 2110.0, "completions/mean_length": 945.26123046875, "completions/mean_terminated_length": 938.2125244140625, "completions/min_length": 372.0, "completions/min_terminated_length": 372.0, "epoch": 0.7220216606498195, "grad_norm": 0.11726341396570206, "kl": 0.0206451416015625, "learning_rate": 1e-06, "loss": 0.0059, "num_tokens": 99224632.0, "reward": 1.1602680683135986, "reward_std": 0.18646904826164246, "rewards/code_format_reward/mean": 0.9754464030265808, "rewards/code_format_reward/std": 0.1549331247806549, "rewards/curriculum_aware_reward_fn/mean": 0.18482144176959991, "rewards/curriculum_aware_reward_fn/std": 0.3438718318939209, "step": 175 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 2001.0, "completions/mean_length": 922.1920166015625, "completions/mean_terminated_length": 915.0917358398438, "completions/min_length": 404.0, "completions/min_terminated_length": 404.0, "epoch": 0.7261474987106756, "grad_norm": 0.11513147503137589, "kl": 0.0187530517578125, "learning_rate": 1e-06, "loss": 0.0205, "num_tokens": 99906651.0, "reward": 1.2103794813156128, "reward_std": 0.19868923723697662, "rewards/code_format_reward/mean": 0.9888392686843872, "rewards/code_format_reward/std": 0.10517053306102753, "rewards/curriculum_aware_reward_fn/mean": 0.2215401828289032, "rewards/curriculum_aware_reward_fn/std": 0.35825085639953613, "step": 176 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1882.0, "completions/max_terminated_length": 1882.0, "completions/mean_length": 927.2656860351562, "completions/mean_terminated_length": 927.2656860351562, "completions/min_length": 379.0, "completions/min_terminated_length": 379.0, "epoch": 0.7302733367715317, "grad_norm": 0.08173976838588715, "kl": 0.020538330078125, "learning_rate": 1e-06, "loss": -0.0051, "num_tokens": 100590446.0, "reward": 1.1069196462631226, "reward_std": 0.10365121811628342, "rewards/code_format_reward/mean": 0.9866071343421936, "rewards/code_format_reward/std": 0.11507844179868698, "rewards/curriculum_aware_reward_fn/mean": 0.12031250447034836, "rewards/curriculum_aware_reward_fn/std": 0.2992539405822754, "step": 177 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1838.0, "completions/max_terminated_length": 1838.0, "completions/mean_length": 869.2455444335938, "completions/mean_terminated_length": 869.2455444335938, "completions/min_length": 380.0, "completions/min_terminated_length": 380.0, "epoch": 0.7343991748323878, "grad_norm": 0.11539430916309357, "kl": 0.0206451416015625, "learning_rate": 1e-06, "loss": -0.0019, "num_tokens": 101260260.0, "reward": 1.226339340209961, "reward_std": 0.15384316444396973, "rewards/code_format_reward/mean": 0.9910714030265808, "rewards/code_format_reward/std": 0.09417349100112915, "rewards/curriculum_aware_reward_fn/mean": 0.23526787757873535, "rewards/curriculum_aware_reward_fn/std": 0.37475648522377014, "step": 178 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2421.0, "completions/max_terminated_length": 2421.0, "completions/mean_length": 875.7455444335938, "completions/mean_terminated_length": 875.7455444335938, "completions/min_length": 296.0, "completions/min_terminated_length": 296.0, "epoch": 0.7385250128932439, "grad_norm": 0.1082959920167923, "kl": 0.0197296142578125, "learning_rate": 1e-06, "loss": 0.0096, "num_tokens": 101901675.0, "reward": 1.2039064168930054, "reward_std": 0.16270264983177185, "rewards/code_format_reward/mean": 0.9955357313156128, "rewards/code_format_reward/std": 0.06674052774906158, "rewards/curriculum_aware_reward_fn/mean": 0.20837053656578064, "rewards/curriculum_aware_reward_fn/std": 0.3540153205394745, "step": 179 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2130.0, "completions/max_terminated_length": 2130.0, "completions/mean_length": 987.4777221679688, "completions/mean_terminated_length": 987.4777221679688, "completions/min_length": 374.0, "completions/min_terminated_length": 374.0, "epoch": 0.7426508509541001, "grad_norm": 0.11390314996242523, "kl": 0.01729583740234375, "learning_rate": 1e-06, "loss": 0.0114, "num_tokens": 102603751.0, "reward": 1.1837053298950195, "reward_std": 0.18950296938419342, "rewards/code_format_reward/mean": 0.9888392686843872, "rewards/code_format_reward/std": 0.10517053306102753, "rewards/curriculum_aware_reward_fn/mean": 0.19486607611179352, "rewards/curriculum_aware_reward_fn/std": 0.31729841232299805, "step": 180 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2083.0, "completions/max_terminated_length": 2083.0, "completions/mean_length": 966.7701416015625, "completions/mean_terminated_length": 966.7701416015625, "completions/min_length": 393.0, "completions/min_terminated_length": 393.0, "epoch": 0.7467766890149562, "grad_norm": 0.12506896257400513, "kl": 0.0182952880859375, "learning_rate": 1e-06, "loss": 0.0052, "num_tokens": 103318418.0, "reward": 1.2953126430511475, "reward_std": 0.21441936492919922, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843171834946, "rewards/curriculum_aware_reward_fn/mean": 0.30200889706611633, "rewards/curriculum_aware_reward_fn/std": 0.38095954060554504, "step": 181 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1847.0, "completions/max_terminated_length": 1847.0, "completions/mean_length": 940.888427734375, "completions/mean_terminated_length": 940.888427734375, "completions/min_length": 353.0, "completions/min_terminated_length": 353.0, "epoch": 0.7509025270758123, "grad_norm": 0.11832232028245926, "kl": 0.021331787109375, "learning_rate": 1e-06, "loss": -0.0078, "num_tokens": 103995100.0, "reward": 1.2311385869979858, "reward_std": 0.21622446179389954, "rewards/code_format_reward/mean": 0.9955357313156128, "rewards/code_format_reward/std": 0.06674052774906158, "rewards/curriculum_aware_reward_fn/mean": 0.23560269176959991, "rewards/curriculum_aware_reward_fn/std": 0.3684559762477875, "step": 182 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1836.0, "completions/max_terminated_length": 1836.0, "completions/mean_length": 899.8928833007812, "completions/mean_terminated_length": 899.8928833007812, "completions/min_length": 331.0, "completions/min_terminated_length": 331.0, "epoch": 0.7550283651366684, "grad_norm": 0.10330358892679214, "kl": 0.0203094482421875, "learning_rate": 1e-06, "loss": 0.0078, "num_tokens": 104657679.0, "reward": 1.1783483028411865, "reward_std": 0.15940025448799133, "rewards/code_format_reward/mean": 0.9888392686843872, "rewards/code_format_reward/std": 0.10517053306102753, "rewards/curriculum_aware_reward_fn/mean": 0.18950892984867096, "rewards/curriculum_aware_reward_fn/std": 0.34344515204429626, "step": 183 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2433.0, "completions/max_terminated_length": 2433.0, "completions/mean_length": 949.3170166015625, "completions/mean_terminated_length": 949.3170166015625, "completions/min_length": 239.0, "completions/min_terminated_length": 239.0, "epoch": 0.7591542031975245, "grad_norm": 0.11373110860586166, "kl": 0.019683837890625, "learning_rate": 1e-06, "loss": 0.007, "num_tokens": 105367656.0, "reward": 1.1863839626312256, "reward_std": 0.13826879858970642, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843916893005, "rewards/curriculum_aware_reward_fn/mean": 0.1930803507566452, "rewards/curriculum_aware_reward_fn/std": 0.3279607594013214, "step": 184 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1803.0, "completions/max_terminated_length": 1803.0, "completions/mean_length": 931.4063110351562, "completions/mean_terminated_length": 931.4063110351562, "completions/min_length": 338.0, "completions/min_terminated_length": 338.0, "epoch": 0.7632800412583806, "grad_norm": 0.11785022914409637, "kl": 0.01873779296875, "learning_rate": 1e-06, "loss": 0.0224, "num_tokens": 106054755.0, "reward": 1.2684152126312256, "reward_std": 0.18439286947250366, "rewards/code_format_reward/mean": 0.9977678656578064, "rewards/code_format_reward/std": 0.047245558351278305, "rewards/curriculum_aware_reward_fn/mean": 0.2706473469734192, "rewards/curriculum_aware_reward_fn/std": 0.3833547532558441, "step": 185 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2402.0, "completions/max_terminated_length": 2402.0, "completions/mean_length": 911.2098388671875, "completions/mean_terminated_length": 911.2098388671875, "completions/min_length": 269.0, "completions/min_terminated_length": 269.0, "epoch": 0.7674058793192368, "grad_norm": 0.11173910647630692, "kl": 0.0190582275390625, "learning_rate": 1e-06, "loss": 0.0121, "num_tokens": 106722999.0, "reward": 1.270535945892334, "reward_std": 0.16392873227596283, "rewards/code_format_reward/mean": 1.0, "rewards/code_format_reward/std": 0.0, "rewards/curriculum_aware_reward_fn/mean": 0.2705357074737549, "rewards/curriculum_aware_reward_fn/std": 0.39146676659584045, "step": 186 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1869.0, "completions/max_terminated_length": 1869.0, "completions/mean_length": 962.2120971679688, "completions/mean_terminated_length": 962.2120971679688, "completions/min_length": 395.0, "completions/min_terminated_length": 395.0, "epoch": 0.7715317173800929, "grad_norm": 0.0817779004573822, "kl": 0.0191192626953125, "learning_rate": 1e-06, "loss": 0.0035, "num_tokens": 107421631.0, "reward": 1.1023437976837158, "reward_std": 0.09970241039991379, "rewards/code_format_reward/mean": 0.9977678656578064, "rewards/code_format_reward/std": 0.047245558351278305, "rewards/curriculum_aware_reward_fn/mean": 0.10457588732242584, "rewards/curriculum_aware_reward_fn/std": 0.2742821276187897, "step": 187 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2072.0, "completions/max_terminated_length": 2072.0, "completions/mean_length": 942.4620971679688, "completions/mean_terminated_length": 942.4620971679688, "completions/min_length": 283.0, "completions/min_terminated_length": 283.0, "epoch": 0.7756575554409489, "grad_norm": 0.11250163614749908, "kl": 0.0191650390625, "learning_rate": 1e-06, "loss": 0.0026, "num_tokens": 108106683.0, "reward": 1.1853796243667603, "reward_std": 0.16808848083019257, "rewards/code_format_reward/mean": 0.9977678656578064, "rewards/code_format_reward/std": 0.047245558351278305, "rewards/curriculum_aware_reward_fn/mean": 0.18761160969734192, "rewards/curriculum_aware_reward_fn/std": 0.31132158637046814, "step": 188 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1706.0, "completions/max_terminated_length": 1706.0, "completions/mean_length": 929.8035888671875, "completions/mean_terminated_length": 929.8035888671875, "completions/min_length": 350.0, "completions/min_terminated_length": 350.0, "epoch": 0.779783393501805, "grad_norm": 0.1267770379781723, "kl": 0.01873779296875, "learning_rate": 1e-06, "loss": 0.0106, "num_tokens": 108781325.0, "reward": 1.305580496788025, "reward_std": 0.2219441533088684, "rewards/code_format_reward/mean": 0.9977678656578064, "rewards/code_format_reward/std": 0.047245558351278305, "rewards/curriculum_aware_reward_fn/mean": 0.30781248211860657, "rewards/curriculum_aware_reward_fn/std": 0.38022324442863464, "step": 189 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2346.0, "completions/max_terminated_length": 2346.0, "completions/mean_length": 914.1652221679688, "completions/mean_terminated_length": 914.1652221679688, "completions/min_length": 349.0, "completions/min_terminated_length": 349.0, "epoch": 0.7839092315626611, "grad_norm": 0.06806844472885132, "kl": 0.0188446044921875, "learning_rate": 1e-06, "loss": 0.0037, "num_tokens": 109443606.0, "reward": 1.0475448369979858, "reward_std": 0.06870421767234802, "rewards/code_format_reward/mean": 0.9955357313156128, "rewards/code_format_reward/std": 0.06674052774906158, "rewards/curriculum_aware_reward_fn/mean": 0.05200892686843872, "rewards/curriculum_aware_reward_fn/std": 0.19431628286838531, "step": 190 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 3219.0, "completions/max_terminated_length": 3219.0, "completions/mean_length": 940.3527221679688, "completions/mean_terminated_length": 940.3527221679688, "completions/min_length": 398.0, "completions/min_terminated_length": 398.0, "epoch": 0.7880350696235173, "grad_norm": 0.11304427683353424, "kl": 0.0201263427734375, "learning_rate": 1e-06, "loss": 0.0053, "num_tokens": 110114014.0, "reward": 1.2850446701049805, "reward_std": 0.23639506101608276, "rewards/code_format_reward/mean": 0.9910714030265808, "rewards/code_format_reward/std": 0.09417349845170975, "rewards/curriculum_aware_reward_fn/mean": 0.2939732074737549, "rewards/curriculum_aware_reward_fn/std": 0.4089702069759369, "step": 191 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2692.0, "completions/max_terminated_length": 2692.0, "completions/mean_length": 973.450927734375, "completions/mean_terminated_length": 973.450927734375, "completions/min_length": 462.0, "completions/min_terminated_length": 462.0, "epoch": 0.7921609076843734, "grad_norm": 0.10210364311933517, "kl": 0.01904296875, "learning_rate": 1e-06, "loss": 0.0145, "num_tokens": 110843133.0, "reward": 1.1609375476837158, "reward_std": 0.1711101233959198, "rewards/code_format_reward/mean": 0.9821428656578064, "rewards/code_format_reward/std": 0.13258016109466553, "rewards/curriculum_aware_reward_fn/mean": 0.17879463732242584, "rewards/curriculum_aware_reward_fn/std": 0.3403628170490265, "step": 192 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.004464285714285698, "completions/max_length": 4096.0, "completions/max_terminated_length": 2787.0, "completions/mean_length": 937.841552734375, "completions/mean_terminated_length": 923.679443359375, "completions/min_length": 320.0, "completions/min_terminated_length": 320.0, "epoch": 0.7962867457452295, "grad_norm": 0.10736164450645447, "kl": 0.0199127197265625, "learning_rate": 1e-06, "loss": 0.0276, "num_tokens": 111536671.0, "reward": 1.187388300895691, "reward_std": 0.1541023552417755, "rewards/code_format_reward/mean": 0.9866071343421936, "rewards/code_format_reward/std": 0.11507843434810638, "rewards/curriculum_aware_reward_fn/mean": 0.20078124105930328, "rewards/curriculum_aware_reward_fn/std": 0.3482690751552582, "step": 193 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2607.0, "completions/max_terminated_length": 2607.0, "completions/mean_length": 946.5178833007812, "completions/mean_terminated_length": 946.5178833007812, "completions/min_length": 360.0, "completions/min_terminated_length": 360.0, "epoch": 0.8004125838060856, "grad_norm": 0.11084363609552383, "kl": 0.020050048828125, "learning_rate": 1e-06, "loss": 0.0217, "num_tokens": 112225101.0, "reward": 1.2250001430511475, "reward_std": 0.17319640517234802, "rewards/code_format_reward/mean": 0.984375, "rewards/code_format_reward/std": 0.12415824085474014, "rewards/curriculum_aware_reward_fn/mean": 0.24062499403953552, "rewards/curriculum_aware_reward_fn/std": 0.37732312083244324, "step": 194 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 2353.0, "completions/mean_length": 897.6719360351562, "completions/mean_terminated_length": 890.5167846679688, "completions/min_length": 211.0, "completions/min_terminated_length": 211.0, "epoch": 0.8045384218669417, "grad_norm": 0.12306180596351624, "kl": 0.0208587646484375, "learning_rate": 1e-06, "loss": -0.0016, "num_tokens": 112886041.0, "reward": 1.243192195892334, "reward_std": 0.1754111498594284, "rewards/code_format_reward/mean": 0.9888392686843872, "rewards/code_format_reward/std": 0.10517053306102753, "rewards/curriculum_aware_reward_fn/mean": 0.2543526589870453, "rewards/curriculum_aware_reward_fn/std": 0.3672681748867035, "step": 195 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 2551.0, "completions/mean_length": 934.138427734375, "completions/mean_terminated_length": 927.0648803710938, "completions/min_length": 385.0, "completions/min_terminated_length": 385.0, "epoch": 0.8086642599277978, "grad_norm": 0.1201872006058693, "kl": 0.018890380859375, "learning_rate": 1e-06, "loss": 0.0257, "num_tokens": 113576023.0, "reward": 1.1931921243667603, "reward_std": 0.18748697638511658, "rewards/code_format_reward/mean": 0.9888392686843872, "rewards/code_format_reward/std": 0.10517053306102753, "rewards/curriculum_aware_reward_fn/mean": 0.20435269176959991, "rewards/curriculum_aware_reward_fn/std": 0.3504360020160675, "step": 196 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1868.0, "completions/max_terminated_length": 1868.0, "completions/mean_length": 932.01123046875, "completions/mean_terminated_length": 932.01123046875, "completions/min_length": 286.0, "completions/min_terminated_length": 286.0, "epoch": 0.812790097988654, "grad_norm": 0.1020113155245781, "kl": 0.02105712890625, "learning_rate": 1e-06, "loss": -0.001, "num_tokens": 114260593.0, "reward": 1.1170759201049805, "reward_std": 0.1418529897928238, "rewards/code_format_reward/mean": 0.984375, "rewards/code_format_reward/std": 0.12415824085474014, "rewards/curriculum_aware_reward_fn/mean": 0.13270089030265808, "rewards/curriculum_aware_reward_fn/std": 0.29727858304977417, "step": 197 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2124.0, "completions/max_terminated_length": 2124.0, "completions/mean_length": 904.7723388671875, "completions/mean_terminated_length": 904.7723388671875, "completions/min_length": 298.0, "completions/min_terminated_length": 298.0, "epoch": 0.8169159360495101, "grad_norm": 0.11794961988925934, "kl": 0.020111083984375, "learning_rate": 1e-06, "loss": 0.0098, "num_tokens": 114919800.0, "reward": 1.2450894117355347, "reward_std": 0.20500408113002777, "rewards/code_format_reward/mean": 0.9910714030265808, "rewards/code_format_reward/std": 0.09417349100112915, "rewards/curriculum_aware_reward_fn/mean": 0.2540178596973419, "rewards/curriculum_aware_reward_fn/std": 0.37923675775527954, "step": 198 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1837.0, "completions/max_terminated_length": 1837.0, "completions/mean_length": 854.372802734375, "completions/mean_terminated_length": 854.372802734375, "completions/min_length": 294.0, "completions/min_terminated_length": 294.0, "epoch": 0.8210417741103662, "grad_norm": 0.1015404462814331, "kl": 0.0208892822265625, "learning_rate": 1e-06, "loss": 0.0094, "num_tokens": 115561699.0, "reward": 1.1597099304199219, "reward_std": 0.12012747675180435, "rewards/code_format_reward/mean": 0.9955357313156128, "rewards/code_format_reward/std": 0.06674052774906158, "rewards/curriculum_aware_reward_fn/mean": 0.16417410969734192, "rewards/curriculum_aware_reward_fn/std": 0.33283236622810364, "step": 199 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 2058.0, "completions/mean_length": 923.69873046875, "completions/mean_terminated_length": 916.601806640625, "completions/min_length": 368.0, "completions/min_terminated_length": 368.0, "epoch": 0.8251676121712223, "grad_norm": 0.1096581444144249, "kl": 0.0187530517578125, "learning_rate": 1e-06, "loss": 0.0114, "num_tokens": 116257476.0, "reward": 1.1924107074737549, "reward_std": 0.17151717841625214, "rewards/code_format_reward/mean": 0.984375, "rewards/code_format_reward/std": 0.12415824085474014, "rewards/curriculum_aware_reward_fn/mean": 0.20803570747375488, "rewards/curriculum_aware_reward_fn/std": 0.3401739001274109, "step": 200 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2622.0, "completions/max_terminated_length": 2622.0, "completions/mean_length": 881.825927734375, "completions/mean_terminated_length": 881.825927734375, "completions/min_length": 364.0, "completions/min_terminated_length": 364.0, "epoch": 0.8292934502320783, "grad_norm": 0.07463487982749939, "kl": 0.0205078125, "learning_rate": 1e-06, "loss": 0.0039, "num_tokens": 116921236.0, "reward": 1.1200892925262451, "reward_std": 0.07546036690473557, "rewards/code_format_reward/mean": 0.9977678656578064, "rewards/code_format_reward/std": 0.047245558351278305, "rewards/curriculum_aware_reward_fn/mean": 0.12232142686843872, "rewards/curriculum_aware_reward_fn/std": 0.3022724688053131, "step": 201 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 1778.0, "completions/mean_length": 880.982177734375, "completions/mean_terminated_length": 873.7897338867188, "completions/min_length": 260.0, "completions/min_terminated_length": 260.0, "epoch": 0.8334192882929345, "grad_norm": 0.09505066275596619, "kl": 0.0218048095703125, "learning_rate": 1e-06, "loss": 0.0163, "num_tokens": 117575133.0, "reward": 1.1426340341567993, "reward_std": 0.1230761781334877, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843171834946, "rewards/curriculum_aware_reward_fn/mean": 0.14933034777641296, "rewards/curriculum_aware_reward_fn/std": 0.30984875559806824, "step": 202 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 1941.0, "completions/mean_length": 938.7589721679688, "completions/mean_terminated_length": 931.6957397460938, "completions/min_length": 366.0, "completions/min_terminated_length": 366.0, "epoch": 0.8375451263537906, "grad_norm": 0.6744564771652222, "kl": 0.1052398681640625, "learning_rate": 1e-06, "loss": 0.0108, "num_tokens": 118277699.0, "reward": 1.0709822177886963, "reward_std": 0.1187218725681305, "rewards/code_format_reward/mean": 0.9866071343421936, "rewards/code_format_reward/std": 0.11507843434810638, "rewards/curriculum_aware_reward_fn/mean": 0.08437498658895493, "rewards/curriculum_aware_reward_fn/std": 0.23442834615707397, "step": 203 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2879.0, "completions/max_terminated_length": 2879.0, "completions/mean_length": 904.99560546875, "completions/mean_terminated_length": 904.99560546875, "completions/min_length": 384.0, "completions/min_terminated_length": 384.0, "epoch": 0.8416709644146467, "grad_norm": 0.10949993878602982, "kl": 0.02069091796875, "learning_rate": 1e-06, "loss": 0.0108, "num_tokens": 118947968.0, "reward": 1.2011160850524902, "reward_std": 0.16616937518119812, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843171834946, "rewards/curriculum_aware_reward_fn/mean": 0.20781250298023224, "rewards/curriculum_aware_reward_fn/std": 0.34411853551864624, "step": 204 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2407.0, "completions/max_terminated_length": 2407.0, "completions/mean_length": 902.8214721679688, "completions/mean_terminated_length": 902.8214721679688, "completions/min_length": 243.0, "completions/min_terminated_length": 243.0, "epoch": 0.8457968024755028, "grad_norm": 0.13115480542182922, "kl": 0.0220184326171875, "learning_rate": 1e-06, "loss": -0.0031, "num_tokens": 119617214.0, "reward": 1.2085938453674316, "reward_std": 0.21638847887516022, "rewards/code_format_reward/mean": 0.9910714030265808, "rewards/code_format_reward/std": 0.09417349100112915, "rewards/curriculum_aware_reward_fn/mean": 0.21752233803272247, "rewards/curriculum_aware_reward_fn/std": 0.3614380657672882, "step": 205 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2522.0, "completions/max_terminated_length": 2522.0, "completions/mean_length": 886.8125610351562, "completions/mean_terminated_length": 886.8125610351562, "completions/min_length": 314.0, "completions/min_terminated_length": 314.0, "epoch": 0.8499226405363589, "grad_norm": 0.11938807368278503, "kl": 0.021240234375, "learning_rate": 1e-06, "loss": -0.0062, "num_tokens": 120275620.0, "reward": 1.205357313156128, "reward_std": 0.2014761120080948, "rewards/code_format_reward/mean": 0.984375, "rewards/code_format_reward/std": 0.12415824085474014, "rewards/curriculum_aware_reward_fn/mean": 0.2209821492433548, "rewards/curriculum_aware_reward_fn/std": 0.35908710956573486, "step": 206 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2088.0, "completions/max_terminated_length": 2088.0, "completions/mean_length": 864.7076416015625, "completions/mean_terminated_length": 864.7076416015625, "completions/min_length": 357.0, "completions/min_terminated_length": 357.0, "epoch": 0.8540484785972151, "grad_norm": 0.17186300456523895, "kl": 0.0281982421875, "learning_rate": 1e-06, "loss": -0.0065, "num_tokens": 120919727.0, "reward": 1.206808090209961, "reward_std": 0.34344661235809326, "rewards/code_format_reward/mean": 0.9017857313156128, "rewards/code_format_reward/std": 0.2979368567466736, "rewards/curriculum_aware_reward_fn/mean": 0.30502229928970337, "rewards/curriculum_aware_reward_fn/std": 0.39650654792785645, "step": 207 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 2096.0, "completions/mean_length": 879.57373046875, "completions/mean_terminated_length": 872.3780517578125, "completions/min_length": 358.0, "completions/min_terminated_length": 358.0, "epoch": 0.8581743166580712, "grad_norm": 0.23083628714084625, "kl": 0.0210723876953125, "learning_rate": 1e-06, "loss": -0.0187, "num_tokens": 121574988.0, "reward": 1.0363839864730835, "reward_std": 0.43614667654037476, "rewards/code_format_reward/mean": 0.796875, "rewards/code_format_reward/std": 0.4027745723724365, "rewards/curriculum_aware_reward_fn/mean": 0.23950894176959991, "rewards/curriculum_aware_reward_fn/std": 0.36418426036834717, "step": 208 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2070.0, "completions/max_terminated_length": 2070.0, "completions/mean_length": 880.3348388671875, "completions/mean_terminated_length": 880.3348388671875, "completions/min_length": 303.0, "completions/min_terminated_length": 303.0, "epoch": 0.8623001547189273, "grad_norm": 0.22123456001281738, "kl": 0.021392822265625, "learning_rate": 1e-06, "loss": -0.0464, "num_tokens": 122248258.0, "reward": 1.0797991752624512, "reward_std": 0.44785869121551514, "rewards/code_format_reward/mean": 0.7790178656578064, "rewards/code_format_reward/std": 0.4153723120689392, "rewards/curriculum_aware_reward_fn/mean": 0.30078125, "rewards/curriculum_aware_reward_fn/std": 0.4084598124027252, "step": 209 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1805.0, "completions/max_terminated_length": 1805.0, "completions/mean_length": 879.57373046875, "completions/mean_terminated_length": 879.57373046875, "completions/min_length": 316.0, "completions/min_terminated_length": 316.0, "epoch": 0.8664259927797834, "grad_norm": 0.201498344540596, "kl": 0.0277252197265625, "learning_rate": 1e-06, "loss": -0.0293, "num_tokens": 122909000.0, "reward": 1.0529018640518188, "reward_std": 0.4085477590560913, "rewards/code_format_reward/mean": 0.8191964030265808, "rewards/code_format_reward/std": 0.38528555631637573, "rewards/curriculum_aware_reward_fn/mean": 0.23370537161827087, "rewards/curriculum_aware_reward_fn/std": 0.3929497003555298, "step": 210 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2125.0, "completions/max_terminated_length": 2125.0, "completions/mean_length": 839.1719360351562, "completions/mean_terminated_length": 839.1719360351562, "completions/min_length": 114.0, "completions/min_terminated_length": 114.0, "epoch": 0.8705518308406395, "grad_norm": 0.16437368094921112, "kl": 0.024566650390625, "learning_rate": 1e-06, "loss": -0.0027, "num_tokens": 123539679.0, "reward": 1.2116072177886963, "reward_std": 0.31230318546295166, "rewards/code_format_reward/mean": 0.9196428656578064, "rewards/code_format_reward/std": 0.2721492052078247, "rewards/curriculum_aware_reward_fn/mean": 0.2919642925262451, "rewards/curriculum_aware_reward_fn/std": 0.4206793010234833, "step": 211 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2157.0, "completions/max_terminated_length": 2157.0, "completions/mean_length": 826.0491333007812, "completions/mean_terminated_length": 826.0491333007812, "completions/min_length": 277.0, "completions/min_terminated_length": 277.0, "epoch": 0.8746776689014956, "grad_norm": 0.13236747682094574, "kl": 0.023529052734375, "learning_rate": 1e-06, "loss": 0.0044, "num_tokens": 124173264.0, "reward": 1.229241132736206, "reward_std": 0.2566564977169037, "rewards/code_format_reward/mean": 0.9665178656578064, "rewards/code_format_reward/std": 0.1800929754972458, "rewards/curriculum_aware_reward_fn/mean": 0.2627232074737549, "rewards/curriculum_aware_reward_fn/std": 0.3902831971645355, "step": 212 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2208.0, "completions/max_terminated_length": 2208.0, "completions/mean_length": 889.0469360351562, "completions/mean_terminated_length": 889.0469360351562, "completions/min_length": 338.0, "completions/min_terminated_length": 338.0, "epoch": 0.8788035069623518, "grad_norm": 0.12648367881774902, "kl": 0.0223846435546875, "learning_rate": 1e-06, "loss": -0.0045, "num_tokens": 124843458.0, "reward": 1.2270090579986572, "reward_std": 0.23483841121196747, "rewards/code_format_reward/mean": 0.9776785969734192, "rewards/code_format_reward/std": 0.14789186418056488, "rewards/curriculum_aware_reward_fn/mean": 0.24933035671710968, "rewards/curriculum_aware_reward_fn/std": 0.3767273426055908, "step": 213 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 2030.0, "completions/mean_length": 891.341552734375, "completions/mean_terminated_length": 884.1722412109375, "completions/min_length": 345.0, "completions/min_terminated_length": 345.0, "epoch": 0.8829293450232079, "grad_norm": 0.13447290658950806, "kl": 0.0208892822265625, "learning_rate": 1e-06, "loss": 0.0214, "num_tokens": 125522884.0, "reward": 1.2450892925262451, "reward_std": 0.25819045305252075, "rewards/code_format_reward/mean": 0.9799107313156128, "rewards/code_format_reward/std": 0.14046260714530945, "rewards/curriculum_aware_reward_fn/mean": 0.2651785910129547, "rewards/curriculum_aware_reward_fn/std": 0.35792961716651917, "step": 214 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 2202.0, "completions/mean_length": 878.7545166015625, "completions/mean_terminated_length": 871.5570678710938, "completions/min_length": 333.0, "completions/min_terminated_length": 333.0, "epoch": 0.887055183084064, "grad_norm": 0.11783410608768463, "kl": 0.02325439453125, "learning_rate": 1e-06, "loss": 0.0183, "num_tokens": 126175748.0, "reward": 1.2014509439468384, "reward_std": 0.20183153450489044, "rewards/code_format_reward/mean": 0.9776785969734192, "rewards/code_format_reward/std": 0.1478918492794037, "rewards/curriculum_aware_reward_fn/mean": 0.2237723171710968, "rewards/curriculum_aware_reward_fn/std": 0.36135873198509216, "step": 215 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1995.0, "completions/max_terminated_length": 1995.0, "completions/mean_length": 842.4754638671875, "completions/mean_terminated_length": 842.4754638671875, "completions/min_length": 286.0, "completions/min_terminated_length": 286.0, "epoch": 0.89118102114492, "grad_norm": 0.13752838969230652, "kl": 0.021209716796875, "learning_rate": 1e-06, "loss": 0.0034, "num_tokens": 126822341.0, "reward": 1.2912946939468384, "reward_std": 0.23720626533031464, "rewards/code_format_reward/mean": 0.9866071343421936, "rewards/code_format_reward/std": 0.11507844179868698, "rewards/curriculum_aware_reward_fn/mean": 0.3046875, "rewards/curriculum_aware_reward_fn/std": 0.4008384943008423, "step": 216 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2195.0, "completions/max_terminated_length": 2195.0, "completions/mean_length": 820.9844360351562, "completions/mean_terminated_length": 820.9844360351562, "completions/min_length": 362.0, "completions/min_terminated_length": 362.0, "epoch": 0.8953068592057761, "grad_norm": 0.13490068912506104, "kl": 0.020477294921875, "learning_rate": 1e-06, "loss": 0.0171, "num_tokens": 127452264.0, "reward": 1.2213170528411865, "reward_std": 0.21540036797523499, "rewards/code_format_reward/mean": 0.984375, "rewards/code_format_reward/std": 0.12415824085474014, "rewards/curriculum_aware_reward_fn/mean": 0.23694197833538055, "rewards/curriculum_aware_reward_fn/std": 0.36264657974243164, "step": 217 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 1837.0, "completions/mean_length": 790.1495971679688, "completions/mean_terminated_length": 782.75390625, "completions/min_length": 267.0, "completions/min_terminated_length": 267.0, "epoch": 0.8994326972666323, "grad_norm": 0.12708254158496857, "kl": 0.023223876953125, "learning_rate": 1e-06, "loss": 0.0098, "num_tokens": 128063648.0, "reward": 1.270647406578064, "reward_std": 0.2100793719291687, "rewards/code_format_reward/mean": 0.9866071343421936, "rewards/code_format_reward/std": 0.11507843434810638, "rewards/curriculum_aware_reward_fn/mean": 0.2840401828289032, "rewards/curriculum_aware_reward_fn/std": 0.40282386541366577, "step": 218 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2020.0, "completions/max_terminated_length": 2020.0, "completions/mean_length": 817.5848388671875, "completions/mean_terminated_length": 817.5848388671875, "completions/min_length": 247.0, "completions/min_terminated_length": 247.0, "epoch": 0.9035585353274884, "grad_norm": 0.1269751638174057, "kl": 0.0226593017578125, "learning_rate": 1e-06, "loss": 0.0034, "num_tokens": 128700273.0, "reward": 1.2515625953674316, "reward_std": 0.1875469833612442, "rewards/code_format_reward/mean": 0.9910714030265808, "rewards/code_format_reward/std": 0.09417349100112915, "rewards/curriculum_aware_reward_fn/mean": 0.2604910731315613, "rewards/curriculum_aware_reward_fn/std": 0.3770267963409424, "step": 219 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2001.0, "completions/max_terminated_length": 2001.0, "completions/mean_length": 776.8795166015625, "completions/mean_terminated_length": 776.8795166015625, "completions/min_length": 276.0, "completions/min_terminated_length": 276.0, "epoch": 0.9076843733883445, "grad_norm": 0.10961435735225677, "kl": 0.0236358642578125, "learning_rate": 1e-06, "loss": 0.0094, "num_tokens": 129302968.0, "reward": 1.1479911804199219, "reward_std": 0.11874659359455109, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843171834946, "rewards/curriculum_aware_reward_fn/mean": 0.15468750894069672, "rewards/curriculum_aware_reward_fn/std": 0.31463250517845154, "step": 220 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2858.0, "completions/max_terminated_length": 2858.0, "completions/mean_length": 794.904052734375, "completions/mean_terminated_length": 794.904052734375, "completions/min_length": 301.0, "completions/min_terminated_length": 301.0, "epoch": 0.9118102114492006, "grad_norm": 0.10039176791906357, "kl": 0.02142333984375, "learning_rate": 1e-06, "loss": 0.0071, "num_tokens": 129910805.0, "reward": 1.1400669813156128, "reward_std": 0.10353995114564896, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843916893005, "rewards/curriculum_aware_reward_fn/mean": 0.1467633992433548, "rewards/curriculum_aware_reward_fn/std": 0.3104727268218994, "step": 221 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2195.0, "completions/max_terminated_length": 2195.0, "completions/mean_length": 822.4620971679688, "completions/mean_terminated_length": 822.4620971679688, "completions/min_length": 245.0, "completions/min_terminated_length": 245.0, "epoch": 0.9159360495100567, "grad_norm": 0.12353157997131348, "kl": 0.0208282470703125, "learning_rate": 1e-06, "loss": 0.0081, "num_tokens": 130542748.0, "reward": 1.1943080425262451, "reward_std": 0.17369195818901062, "rewards/code_format_reward/mean": 0.9910714030265808, "rewards/code_format_reward/std": 0.09417349100112915, "rewards/curriculum_aware_reward_fn/mean": 0.20323659479618073, "rewards/curriculum_aware_reward_fn/std": 0.35851049423217773, "step": 222 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2351.0, "completions/max_terminated_length": 2351.0, "completions/mean_length": 870.6406860351562, "completions/mean_terminated_length": 870.6406860351562, "completions/min_length": 300.0, "completions/min_terminated_length": 300.0, "epoch": 0.9200618875709129, "grad_norm": 0.10986126214265823, "kl": 0.020416259765625, "learning_rate": 1e-06, "loss": -0.0043, "num_tokens": 131197990.0, "reward": 1.2091518640518188, "reward_std": 0.1457216739654541, "rewards/code_format_reward/mean": 0.9910714030265808, "rewards/code_format_reward/std": 0.09417349100112915, "rewards/curriculum_aware_reward_fn/mean": 0.21808035671710968, "rewards/curriculum_aware_reward_fn/std": 0.40909281373023987, "step": 223 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1816.0, "completions/max_terminated_length": 1816.0, "completions/mean_length": 869.6272583007812, "completions/mean_terminated_length": 869.6272583007812, "completions/min_length": 340.0, "completions/min_terminated_length": 340.0, "epoch": 0.924187725631769, "grad_norm": 0.12486883997917175, "kl": 0.02044677734375, "learning_rate": 1e-06, "loss": -0.0037, "num_tokens": 131865469.0, "reward": 1.2027902603149414, "reward_std": 0.18685774505138397, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843171834946, "rewards/curriculum_aware_reward_fn/mean": 0.20948660373687744, "rewards/curriculum_aware_reward_fn/std": 0.3364081084728241, "step": 224 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2009.0, "completions/max_terminated_length": 2009.0, "completions/mean_length": 806.1607666015625, "completions/mean_terminated_length": 806.1607666015625, "completions/min_length": 286.0, "completions/min_terminated_length": 286.0, "epoch": 0.9283135636926251, "grad_norm": 0.10445675998926163, "kl": 0.0204010009765625, "learning_rate": 1e-06, "loss": -0.0009, "num_tokens": 132495247.0, "reward": 1.1902902126312256, "reward_std": 0.1378038227558136, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843171834946, "rewards/curriculum_aware_reward_fn/mean": 0.1969866007566452, "rewards/curriculum_aware_reward_fn/std": 0.36033302545547485, "step": 225 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1777.0, "completions/max_terminated_length": 1777.0, "completions/mean_length": 823.3705444335938, "completions/mean_terminated_length": 823.3705444335938, "completions/min_length": 234.0, "completions/min_terminated_length": 234.0, "epoch": 0.9324394017534812, "grad_norm": 0.1085299551486969, "kl": 0.020965576171875, "learning_rate": 1e-06, "loss": 0.001, "num_tokens": 133127702.0, "reward": 1.165178656578064, "reward_std": 0.14776389300823212, "rewards/code_format_reward/mean": 0.9910714030265808, "rewards/code_format_reward/std": 0.09417349100112915, "rewards/curriculum_aware_reward_fn/mean": 0.1741071492433548, "rewards/curriculum_aware_reward_fn/std": 0.3225827217102051, "step": 226 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2451.0, "completions/max_terminated_length": 2451.0, "completions/mean_length": 804.1317138671875, "completions/mean_terminated_length": 804.1317138671875, "completions/min_length": 276.0, "completions/min_terminated_length": 276.0, "epoch": 0.9365652398143373, "grad_norm": 0.1200730949640274, "kl": 0.0230712890625, "learning_rate": 1e-06, "loss": 0.0054, "num_tokens": 133750586.0, "reward": 1.1993305683135986, "reward_std": 0.1256466507911682, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843916893005, "rewards/curriculum_aware_reward_fn/mean": 0.20602677762508392, "rewards/curriculum_aware_reward_fn/std": 0.36172813177108765, "step": 227 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1768.0, "completions/max_terminated_length": 1768.0, "completions/mean_length": 823.8660888671875, "completions/mean_terminated_length": 823.8660888671875, "completions/min_length": 270.0, "completions/min_terminated_length": 270.0, "epoch": 0.9406910778751933, "grad_norm": 0.12881457805633545, "kl": 0.0206298828125, "learning_rate": 1e-06, "loss": -0.0022, "num_tokens": 134384761.0, "reward": 1.2723214626312256, "reward_std": 0.22976413369178772, "rewards/code_format_reward/mean": 0.9888392686843872, "rewards/code_format_reward/std": 0.10517053306102753, "rewards/curriculum_aware_reward_fn/mean": 0.2834821343421936, "rewards/curriculum_aware_reward_fn/std": 0.4023776352405548, "step": 228 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 2208.0, "completions/mean_length": 846.1964721679688, "completions/mean_terminated_length": 838.9262084960938, "completions/min_length": 243.0, "completions/min_terminated_length": 243.0, "epoch": 0.9448169159360496, "grad_norm": 0.13712099194526672, "kl": 0.0245819091796875, "learning_rate": 1e-06, "loss": 0.0112, "num_tokens": 135044956.0, "reward": 1.2349331378936768, "reward_std": 0.20659008622169495, "rewards/code_format_reward/mean": 0.984375, "rewards/code_format_reward/std": 0.12415824085474014, "rewards/curriculum_aware_reward_fn/mean": 0.2505580484867096, "rewards/curriculum_aware_reward_fn/std": 0.3524835407733917, "step": 229 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2086.0, "completions/max_terminated_length": 2086.0, "completions/mean_length": 805.232177734375, "completions/mean_terminated_length": 805.232177734375, "completions/min_length": 270.0, "completions/min_terminated_length": 270.0, "epoch": 0.9489427539969056, "grad_norm": 0.13029730319976807, "kl": 0.0221710205078125, "learning_rate": 1e-06, "loss": -0.0049, "num_tokens": 135675102.0, "reward": 1.2138394117355347, "reward_std": 0.19884708523750305, "rewards/code_format_reward/mean": 0.9821428656578064, "rewards/code_format_reward/std": 0.13258016109466553, "rewards/curriculum_aware_reward_fn/mean": 0.23169644176959991, "rewards/curriculum_aware_reward_fn/std": 0.3652246594429016, "step": 230 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 3023.0, "completions/max_terminated_length": 3023.0, "completions/mean_length": 842.0156860351562, "completions/mean_terminated_length": 842.0156860351562, "completions/min_length": 335.0, "completions/min_terminated_length": 335.0, "epoch": 0.9530685920577617, "grad_norm": 0.14022080600261688, "kl": 0.0362701416015625, "learning_rate": 1e-06, "loss": 0.0177, "num_tokens": 136313957.0, "reward": 1.2387278079986572, "reward_std": 0.1900922954082489, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843916893005, "rewards/curriculum_aware_reward_fn/mean": 0.24542410671710968, "rewards/curriculum_aware_reward_fn/std": 0.37329214811325073, "step": 231 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1625.0, "completions/max_terminated_length": 1625.0, "completions/mean_length": 798.0201416015625, "completions/mean_terminated_length": 798.0201416015625, "completions/min_length": 301.0, "completions/min_terminated_length": 301.0, "epoch": 0.9571944301186178, "grad_norm": 0.1330692321062088, "kl": 0.0215301513671875, "learning_rate": 1e-06, "loss": 0.0001, "num_tokens": 136938177.0, "reward": 1.3068081140518188, "reward_std": 0.21471752226352692, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843916893005, "rewards/curriculum_aware_reward_fn/mean": 0.31350448727607727, "rewards/curriculum_aware_reward_fn/std": 0.40733155608177185, "step": 232 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2625.0, "completions/max_terminated_length": 2625.0, "completions/mean_length": 821.3683471679688, "completions/mean_terminated_length": 821.3683471679688, "completions/min_length": 278.0, "completions/min_terminated_length": 278.0, "epoch": 0.9613202681794739, "grad_norm": 0.1162538155913353, "kl": 0.01971435546875, "learning_rate": 1e-06, "loss": 0.0022, "num_tokens": 137573949.0, "reward": 1.1732144355773926, "reward_std": 0.17516572773456573, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843171834946, "rewards/curriculum_aware_reward_fn/mean": 0.17991070449352264, "rewards/curriculum_aware_reward_fn/std": 0.33746057748794556, "step": 233 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1847.0, "completions/max_terminated_length": 1847.0, "completions/mean_length": 792.8080444335938, "completions/mean_terminated_length": 792.8080444335938, "completions/min_length": 291.0, "completions/min_terminated_length": 291.0, "epoch": 0.9654461062403301, "grad_norm": 0.14185446500778198, "kl": 0.0202789306640625, "learning_rate": 1e-06, "loss": 0.0265, "num_tokens": 138201084.0, "reward": 1.223772406578064, "reward_std": 0.20426253974437714, "rewards/code_format_reward/mean": 0.9910714030265808, "rewards/code_format_reward/std": 0.09417349100112915, "rewards/curriculum_aware_reward_fn/mean": 0.2327008992433548, "rewards/curriculum_aware_reward_fn/std": 0.35288336873054504, "step": 234 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2577.0, "completions/max_terminated_length": 2577.0, "completions/mean_length": 847.6339721679688, "completions/mean_terminated_length": 847.6339721679688, "completions/min_length": 297.0, "completions/min_terminated_length": 297.0, "epoch": 0.9695719443011862, "grad_norm": 0.1119653508067131, "kl": 0.0201263427734375, "learning_rate": 1e-06, "loss": -0.0115, "num_tokens": 138851794.0, "reward": 1.2417410612106323, "reward_std": 0.15232668817043304, "rewards/code_format_reward/mean": 0.9910714030265808, "rewards/code_format_reward/std": 0.09417349100112915, "rewards/curriculum_aware_reward_fn/mean": 0.2506696283817291, "rewards/curriculum_aware_reward_fn/std": 0.3951603174209595, "step": 235 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1641.0, "completions/max_terminated_length": 1641.0, "completions/mean_length": 788.950927734375, "completions/mean_terminated_length": 788.950927734375, "completions/min_length": 317.0, "completions/min_terminated_length": 317.0, "epoch": 0.9736977823620423, "grad_norm": 0.13373960554599762, "kl": 0.0213165283203125, "learning_rate": 1e-06, "loss": 0.0109, "num_tokens": 139479700.0, "reward": 1.2359375953674316, "reward_std": 0.19174519181251526, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843171834946, "rewards/curriculum_aware_reward_fn/mean": 0.24263392388820648, "rewards/curriculum_aware_reward_fn/std": 0.3762992024421692, "step": 236 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0022321428571429047, "completions/max_length": 4096.0, "completions/max_terminated_length": 2174.0, "completions/mean_length": 805.9620971679688, "completions/mean_terminated_length": 798.601806640625, "completions/min_length": 304.0, "completions/min_terminated_length": 304.0, "epoch": 0.9778236204228984, "grad_norm": 0.13458861410617828, "kl": 0.0209503173828125, "learning_rate": 1e-06, "loss": 0.0251, "num_tokens": 140097055.0, "reward": 1.2815849781036377, "reward_std": 0.23763927817344666, "rewards/code_format_reward/mean": 0.9821428656578064, "rewards/code_format_reward/std": 0.13258016109466553, "rewards/curriculum_aware_reward_fn/mean": 0.29944199323654175, "rewards/curriculum_aware_reward_fn/std": 0.37762463092803955, "step": 237 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2724.0, "completions/max_terminated_length": 2724.0, "completions/mean_length": 798.7455444335938, "completions/mean_terminated_length": 798.7455444335938, "completions/min_length": 377.0, "completions/min_terminated_length": 377.0, "epoch": 0.9819494584837545, "grad_norm": 0.1388353556394577, "kl": 0.02252197265625, "learning_rate": 1e-06, "loss": 0.0075, "num_tokens": 140730847.0, "reward": 1.2100446224212646, "reward_std": 0.22694040834903717, "rewards/code_format_reward/mean": 0.9866071343421936, "rewards/code_format_reward/std": 0.11507843434810638, "rewards/curriculum_aware_reward_fn/mean": 0.22343750298023224, "rewards/curriculum_aware_reward_fn/std": 0.3409222364425659, "step": 238 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2872.0, "completions/max_terminated_length": 2872.0, "completions/mean_length": 823.138427734375, "completions/mean_terminated_length": 823.138427734375, "completions/min_length": 216.0, "completions/min_terminated_length": 216.0, "epoch": 0.9860752965446106, "grad_norm": 0.13200150430202484, "kl": 0.022369384765625, "learning_rate": 1e-06, "loss": 0.0027, "num_tokens": 141366866.0, "reward": 1.252343773841858, "reward_std": 0.20880478620529175, "rewards/code_format_reward/mean": 0.9866071343421936, "rewards/code_format_reward/std": 0.11507843434810638, "rewards/curriculum_aware_reward_fn/mean": 0.2657366096973419, "rewards/curriculum_aware_reward_fn/std": 0.38834115862846375, "step": 239 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 1728.0, "completions/max_terminated_length": 1728.0, "completions/mean_length": 818.9174194335938, "completions/mean_terminated_length": 818.9174194335938, "completions/min_length": 296.0, "completions/min_terminated_length": 296.0, "epoch": 0.9902011346054668, "grad_norm": 0.09640368819236755, "kl": 0.0225830078125, "learning_rate": 1e-06, "loss": 0.0017, "num_tokens": 142006496.0, "reward": 1.14051353931427, "reward_std": 0.0970856174826622, "rewards/code_format_reward/mean": 0.9933035969734192, "rewards/code_format_reward/std": 0.08164843916893005, "rewards/curriculum_aware_reward_fn/mean": 0.14720980823040009, "rewards/curriculum_aware_reward_fn/std": 0.3216255009174347, "step": 240 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 3333.0, "completions/max_terminated_length": 3333.0, "completions/mean_length": 815.5201416015625, "completions/mean_terminated_length": 815.5201416015625, "completions/min_length": 174.0, "completions/min_terminated_length": 174.0, "epoch": 0.9943269726663229, "grad_norm": 0.14960746467113495, "kl": 0.0247039794921875, "learning_rate": 1e-06, "loss": 0.0004, "num_tokens": 142648139.0, "reward": 1.2400671243667603, "reward_std": 0.21601368486881256, "rewards/code_format_reward/mean": 0.9888392686843872, "rewards/code_format_reward/std": 0.10517053306102753, "rewards/curriculum_aware_reward_fn/mean": 0.2512276768684387, "rewards/curriculum_aware_reward_fn/std": 0.37283849716186523, "step": 241 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 2014.0, "completions/max_terminated_length": 2014.0, "completions/mean_length": 856.6094970703125, "completions/mean_terminated_length": 856.6094970703125, "completions/min_length": 321.0, "completions/min_terminated_length": 321.0, "epoch": 0.998452810727179, "grad_norm": 0.11549442261457443, "kl": 0.02239990234375, "learning_rate": 1e-06, "loss": 0.0012, "num_tokens": 143306525.0, "reward": 1.1822545528411865, "reward_std": 0.177961528301239, "rewards/code_format_reward/mean": 0.9910714030265808, "rewards/code_format_reward/std": 0.09417349100112915, "rewards/curriculum_aware_reward_fn/mean": 0.19118304550647736, "rewards/curriculum_aware_reward_fn/std": 0.3384980261325836, "step": 242 }, { "epoch": 0.998452810727179, "step": 242, "total_flos": 0.0, "train_loss": 0.00999828894322686, "train_runtime": 47323.5256, "train_samples_per_second": 0.328, "train_steps_per_second": 0.005 } ], "logging_steps": 1, "max_steps": 242, "num_input_tokens_seen": 143306525, "num_train_epochs": 1, "save_steps": 500, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 0.0, "train_batch_size": 8, "trial_name": null, "trial_params": null }