{ "best_global_step": null, "best_metric": null, "best_model_checkpoint": null, "epoch": 0.4557868815760018, "eval_steps": 500, "global_step": 509, "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": -7.0, "completions/max_length": 1695.0, "completions/max_terminated_length": 1695.0, "completions/mean_length": 590.9921875, "completions/mean_terminated_length": 590.9921875, "completions/min_length": 212.0, "completions/min_terminated_length": 212.0, "epoch": 0.0008954555630176853, "frac_reward_zero_std": 0.03125, "grad_norm": 1.324537021697667, "kl": 0.0518798828125, "learning_rate": 0.0, "loss": 0.0165, "num_tokens": 614764.0, "reward": 0.05976562947034836, "reward_std": 0.0456097275018692, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.59765625, "rewards/format_reward/std": 0.4908501207828522, "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": -7.0, "completions/max_length": 1975.0, "completions/max_terminated_length": 1975.0, "completions/mean_length": 614.201171875, "completions/mean_terminated_length": 614.201171875, "completions/min_length": 144.0, "completions/min_terminated_length": 144.0, "epoch": 0.0017909111260353706, "frac_reward_zero_std": 0.03125, "grad_norm": 1.1898987030436239, "kl": 0.0516357421875, "learning_rate": 3.3333333333333335e-07, "loss": 0.0218, "num_tokens": 1251315.0, "reward": 0.06015624850988388, "reward_std": 0.04386558383703232, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.6015625, "rewards/format_reward/std": 0.4900552034378052, "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1477.0, "completions/mean_length": 619.98828125, "completions/mean_terminated_length": 617.1937255859375, "completions/min_length": 169.0, "completions/min_terminated_length": 169.0, "epoch": 0.002686366689053056, "frac_reward_zero_std": 0.0, "grad_norm": 1.161155142512, "kl": 0.05609130859375, "learning_rate": 6.666666666666667e-07, "loss": 0.0261, "num_tokens": 1895949.0, "reward": 0.06425781548023224, "reward_std": 0.04574279859662056, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.642578125, "rewards/format_reward/std": 0.4797092080116272, "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": -6.9375, "completions/max_length": 2048.0, "completions/max_terminated_length": 2034.0, "completions/mean_length": 641.892578125, "completions/mean_terminated_length": 632.3051147460938, "completions/min_length": 244.0, "completions/min_terminated_length": 244.0, "epoch": 0.003581822252070741, "frac_reward_zero_std": 0.0, "grad_norm": 6.062290227835526, "kl": 0.15771484375, "learning_rate": 1.0000000000000002e-06, "loss": 0.0292, "num_tokens": 2549926.0, "reward": 0.06679688394069672, "reward_std": 0.04454650729894638, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.66796875, "rewards/format_reward/std": 0.47140273451805115, "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1832.0, "completions/mean_length": 602.048828125, "completions/mean_terminated_length": 598.4530029296875, "completions/min_length": 271.0, "completions/min_terminated_length": 271.0, "epoch": 0.0044772778150884264, "frac_reward_zero_std": 0.0625, "grad_norm": 1.0733703140487003, "kl": 0.07025146484375, "learning_rate": 1.3333333333333334e-06, "loss": 0.0339, "num_tokens": 3140687.0, "reward": 0.07949218899011612, "reward_std": 0.03763638436794281, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.794921875, "rewards/format_reward/std": 0.4041535556316376, "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1348.0, "completions/mean_length": 593.43359375, "completions/mean_terminated_length": 590.5870971679688, "completions/min_length": 201.0, "completions/min_terminated_length": 201.0, "epoch": 0.005372733378106112, "frac_reward_zero_std": 0.15625, "grad_norm": 0.46792041518870797, "kl": 0.05133056640625, "learning_rate": 1.6666666666666667e-06, "loss": 0.0476, "num_tokens": 3759885.0, "reward": 0.08847656846046448, "reward_std": 0.028867140412330627, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.884765625, "rewards/format_reward/std": 0.3196168541908264, "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1760.0, "completions/mean_length": 622.640625, "completions/mean_terminated_length": 619.8512573242188, "completions/min_length": 148.0, "completions/min_terminated_length": 148.0, "epoch": 0.006268188941123797, "frac_reward_zero_std": 0.3125, "grad_norm": 0.3715829196297699, "kl": 0.0545654296875, "learning_rate": 2.0000000000000003e-06, "loss": 0.0305, "num_tokens": 4475749.0, "reward": 0.09160156548023224, "reward_std": 0.022055473178625107, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.916015625, "rewards/format_reward/std": 0.2776356339454651, "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": -7.0, "completions/max_length": 1935.0, "completions/max_terminated_length": 1935.0, "completions/mean_length": 581.716796875, "completions/mean_terminated_length": 581.716796875, "completions/min_length": 170.0, "completions/min_terminated_length": 170.0, "epoch": 0.007163644504141482, "frac_reward_zero_std": 0.375, "grad_norm": 0.31874462958935845, "kl": 0.05224609375, "learning_rate": 2.3333333333333336e-06, "loss": 0.0226, "num_tokens": 5085092.0, "reward": 0.09453125298023224, "reward_std": 0.017671994864940643, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.9453125, "rewards/format_reward/std": 0.2275916188955307, "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1669.0, "completions/mean_length": 538.501953125, "completions/mean_terminated_length": 535.5479125976562, "completions/min_length": 194.0, "completions/min_terminated_length": 194.0, "epoch": 0.008059100067159167, "frac_reward_zero_std": 0.71875, "grad_norm": 0.22707584816837714, "kl": 0.05548095703125, "learning_rate": 2.666666666666667e-06, "loss": 0.0184, "num_tokens": 5653829.0, "reward": 0.09804687649011612, "reward_std": 0.007317390292882919, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.98046875, "rewards/format_reward/std": 0.1385180652141571, "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": -7.0, "completions/max_length": 1623.0, "completions/max_terminated_length": 1623.0, "completions/mean_length": 508.15625, "completions/mean_terminated_length": 508.15625, "completions/min_length": 116.0, "completions/min_terminated_length": 116.0, "epoch": 0.008954555630176853, "frac_reward_zero_std": 0.9375, "grad_norm": 0.10154622002931386, "kl": 0.0634765625, "learning_rate": 3e-06, "loss": 0.0023, "num_tokens": 6199173.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1765.0, "completions/mean_length": 553.5234375, "completions/mean_terminated_length": 549.9627685546875, "completions/min_length": 140.0, "completions/min_terminated_length": 140.0, "epoch": 0.009850011193194537, "frac_reward_zero_std": 0.90625, "grad_norm": 0.14394814844886103, "kl": 0.06744384765625, "learning_rate": 3.3333333333333333e-06, "loss": 0.0189, "num_tokens": 6803009.0, "reward": 0.09941406548023224, "reward_std": 0.0023437500931322575, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.994140625, "rewards/format_reward/std": 0.07639661431312561, "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": -7.0, "completions/max_length": 1550.0, "completions/max_terminated_length": 1550.0, "completions/mean_length": 454.251953125, "completions/mean_terminated_length": 454.251953125, "completions/min_length": 78.0, "completions/min_terminated_length": 78.0, "epoch": 0.010745466756212223, "frac_reward_zero_std": 0.9375, "grad_norm": 0.16220747501541533, "kl": 0.0635986328125, "learning_rate": 3.6666666666666666e-06, "loss": 0.0039, "num_tokens": 7340802.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "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": -7.0, "completions/max_length": 1469.0, "completions/max_terminated_length": 1469.0, "completions/mean_length": 469.560546875, "completions/mean_terminated_length": 469.560546875, "completions/min_length": 151.0, "completions/min_terminated_length": 151.0, "epoch": 0.011640922319229908, "frac_reward_zero_std": 0.9375, "grad_norm": 0.1336388368219293, "kl": 0.0775146484375, "learning_rate": 4.000000000000001e-06, "loss": -0.0062, "num_tokens": 7896049.0, "reward": 0.09941406548023224, "reward_std": 0.0018486406188458204, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.994140625, "rewards/format_reward/std": 0.07639661431312561, "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1832.0, "completions/mean_length": 472.15234375, "completions/mean_terminated_length": 469.0684814453125, "completions/min_length": 100.0, "completions/min_terminated_length": 100.0, "epoch": 0.012536377882247594, "frac_reward_zero_std": 0.84375, "grad_norm": 0.2405259644807092, "kl": 0.097412109375, "learning_rate": 4.333333333333334e-06, "loss": -0.0011, "num_tokens": 8492799.0, "reward": 0.09882812947034836, "reward_std": 0.004192390479147434, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.98828125, "rewards/format_reward/std": 0.10772226005792618, "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": -6.875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1911.0, "completions/mean_length": 452.572265625, "completions/mean_terminated_length": 433.5972595214844, "completions/min_length": 98.0, "completions/min_terminated_length": 98.0, "epoch": 0.013431833445265278, "frac_reward_zero_std": 1.0, "grad_norm": 0.1753416604203705, "kl": 0.171630859375, "learning_rate": 4.666666666666667e-06, "loss": 0.0017, "num_tokens": 9043940.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.921875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1929.0, "completions/mean_length": 476.982421875, "completions/mean_terminated_length": 464.00592041015625, "completions/min_length": 121.0, "completions/min_terminated_length": 121.0, "epoch": 0.014327289008282965, "frac_reward_zero_std": 0.96875, "grad_norm": 1.0552577658632907, "kl": 0.169189453125, "learning_rate": 5e-06, "loss": 0.0024, "num_tokens": 9613259.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.90625, "completions/max_length": 2048.0, "completions/max_terminated_length": 1642.0, "completions/mean_length": 417.689453125, "completions/mean_terminated_length": 406.4466552734375, "completions/min_length": 57.0, "completions/min_terminated_length": 57.0, "epoch": 0.015222744571300649, "frac_reward_zero_std": 0.96875, "grad_norm": 0.26917824685344643, "kl": 0.1390380859375, "learning_rate": 4.999952797253148e-06, "loss": 0.0027, "num_tokens": 10112252.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.953125, "completions/max_length": 2048.0, "completions/max_terminated_length": 1843.0, "completions/mean_length": 462.291015625, "completions/mean_terminated_length": 452.94500732421875, "completions/min_length": 110.0, "completions/min_terminated_length": 110.0, "epoch": 0.016118200134318333, "frac_reward_zero_std": 1.0, "grad_norm": 0.2947535239705964, "kl": 0.117431640625, "learning_rate": 4.9998111909931225e-06, "loss": 0.0012, "num_tokens": 10691873.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -7.0, "completions/max_length": 901.0, "completions/max_terminated_length": 901.0, "completions/mean_length": 376.89453125, "completions/mean_terminated_length": 376.89453125, "completions/min_length": 114.0, "completions/min_terminated_length": 114.0, "epoch": 0.01701365569733602, "frac_reward_zero_std": 0.90625, "grad_norm": 0.21872537582484822, "kl": 0.0916748046875, "learning_rate": 4.999575187161439e-06, "loss": 0.0064, "num_tokens": 11171243.0, "reward": 0.09941406548023224, "reward_std": 0.0023437500931322575, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.994140625, "rewards/format_reward/std": 0.07639661431312561, "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": -7.0, "completions/max_length": 1383.0, "completions/max_terminated_length": 1383.0, "completions/mean_length": 386.00390625, "completions/mean_terminated_length": 386.00390625, "completions/min_length": 72.0, "completions/min_terminated_length": 72.0, "epoch": 0.017909111260353706, "frac_reward_zero_std": 0.9375, "grad_norm": 0.13652379466986567, "kl": 0.0869140625, "learning_rate": 4.9992447956603455e-06, "loss": -0.0022, "num_tokens": 11662493.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "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": -7.0, "completions/max_length": 893.0, "completions/max_terminated_length": 893.0, "completions/mean_length": 333.017578125, "completions/mean_terminated_length": 333.017578125, "completions/min_length": 70.0, "completions/min_terminated_length": 70.0, "epoch": 0.01880456682337139, "frac_reward_zero_std": 1.0, "grad_norm": 0.04417150040050226, "kl": 0.095703125, "learning_rate": 4.998820030352409e-06, "loss": 0.001, "num_tokens": 12113686.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -7.0, "completions/max_length": 1031.0, "completions/max_terminated_length": 1031.0, "completions/mean_length": 387.27734375, "completions/mean_terminated_length": 387.27734375, "completions/min_length": 117.0, "completions/min_terminated_length": 117.0, "epoch": 0.019700022386389075, "frac_reward_zero_std": 0.90625, "grad_norm": 0.19366315929923666, "kl": 0.0830078125, "learning_rate": 4.998300909059929e-06, "loss": 0.0035, "num_tokens": 12608996.0, "reward": 0.09902343899011612, "reward_std": 0.0028222277760505676, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.990234375, "rewards/format_reward/std": 0.09843364357948303, "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": -7.0, "completions/max_length": 900.0, "completions/max_terminated_length": 900.0, "completions/mean_length": 393.1796875, "completions/mean_terminated_length": 393.1796875, "completions/min_length": 113.0, "completions/min_terminated_length": 113.0, "epoch": 0.02059547794940676, "frac_reward_zero_std": 0.96875, "grad_norm": 0.08926508511842418, "kl": 0.081298828125, "learning_rate": 4.997687453564198e-06, "loss": 0.0004, "num_tokens": 13080752.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -7.0, "completions/max_length": 1882.0, "completions/max_terminated_length": 1882.0, "completions/mean_length": 444.2734375, "completions/mean_terminated_length": 444.2734375, "completions/min_length": 154.0, "completions/min_terminated_length": 154.0, "epoch": 0.021490933512424447, "frac_reward_zero_std": 0.90625, "grad_norm": 0.12692069614371276, "kl": 0.085693359375, "learning_rate": 4.9969796896045775e-06, "loss": 0.0076, "num_tokens": 13661468.0, "reward": 0.09941406548023224, "reward_std": 0.0023437500931322575, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.994140625, "rewards/format_reward/std": 0.07639661431312561, "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": -6.96875, "completions/max_length": 1259.0, "completions/max_terminated_length": 1259.0, "completions/mean_length": 439.455078125, "completions/mean_terminated_length": 436.3902282714844, "completions/min_length": 160.0, "completions/min_terminated_length": 160.0, "epoch": 0.02238638907544213, "frac_reward_zero_std": 0.90625, "grad_norm": 20.877169149767465, "kl": 0.09375, "learning_rate": 4.996177646877426e-06, "loss": 0.0144, "num_tokens": 14154069.0, "reward": 0.09941406548023224, "reward_std": 0.0023437500931322575, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.994140625, "rewards/format_reward/std": 0.07639661431312561, "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": -6.90625, "completions/max_length": 2048.0, "completions/max_terminated_length": 1895.0, "completions/mean_length": 474.58984375, "completions/mean_terminated_length": 462.3340148925781, "completions/min_length": 135.0, "completions/min_terminated_length": 135.0, "epoch": 0.023281844638459816, "frac_reward_zero_std": 0.90625, "grad_norm": 92.43206042257317, "kl": 7.2659912109375, "learning_rate": 4.995281359034851e-06, "loss": 0.0913, "num_tokens": 14701171.0, "reward": 0.09941406548023224, "reward_std": 0.0023437500931322575, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.994140625, "rewards/format_reward/std": 0.07639661431312561, "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": -6.984375, "completions/max_length": 1354.0, "completions/max_terminated_length": 1354.0, "completions/mean_length": 451.283203125, "completions/mean_terminated_length": 449.9921569824219, "completions/min_length": 125.0, "completions/min_terminated_length": 125.0, "epoch": 0.0241773002014775, "frac_reward_zero_std": 0.9375, "grad_norm": 0.15669283865685832, "kl": 0.0904541015625, "learning_rate": 4.994290863683296e-06, "loss": 0.0103, "num_tokens": 15202436.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "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": -6.984375, "completions/max_length": 1410.0, "completions/max_terminated_length": 1410.0, "completions/mean_length": 478.384765625, "completions/mean_terminated_length": 476.5616455078125, "completions/min_length": 147.0, "completions/min_terminated_length": 147.0, "epoch": 0.025072755764495188, "frac_reward_zero_std": 0.96875, "grad_norm": 0.07431876710651732, "kl": 0.088623046875, "learning_rate": 4.99320620238196e-06, "loss": 0.0051, "num_tokens": 15732137.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.953125, "completions/max_length": 1981.0, "completions/max_terminated_length": 1981.0, "completions/mean_length": 495.1640625, "completions/mean_terminated_length": 489.8369445800781, "completions/min_length": 170.0, "completions/min_terminated_length": 170.0, "epoch": 0.025968211327512872, "frac_reward_zero_std": 0.96875, "grad_norm": 1.419703912441118, "kl": 0.1038818359375, "learning_rate": 4.99202742064106e-06, "loss": -0.0017, "num_tokens": 16339613.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.953125, "completions/max_length": 1830.0, "completions/max_terminated_length": 1830.0, "completions/mean_length": 483.849609375, "completions/mean_terminated_length": 479.77606201171875, "completions/min_length": 151.0, "completions/min_terminated_length": 151.0, "epoch": 0.026863666890530557, "frac_reward_zero_std": 0.96875, "grad_norm": 20.526980124430374, "kl": 0.110595703125, "learning_rate": 4.990754567919917e-06, "loss": 0.0162, "num_tokens": 16889936.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.984375, "completions/max_length": 1573.0, "completions/max_terminated_length": 1573.0, "completions/mean_length": 477.513671875, "completions/mean_terminated_length": 475.369873046875, "completions/min_length": 159.0, "completions/min_terminated_length": 159.0, "epoch": 0.02775912245354824, "frac_reward_zero_std": 1.0, "grad_norm": 0.3115792345317698, "kl": 0.1025390625, "learning_rate": 4.989387697624881e-06, "loss": 0.001, "num_tokens": 17437175.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.984375, "completions/max_length": 1254.0, "completions/max_terminated_length": 1254.0, "completions/mean_length": 497.666015625, "completions/mean_terminated_length": 496.96673583984375, "completions/min_length": 151.0, "completions/min_terminated_length": 151.0, "epoch": 0.02865457801656593, "frac_reward_zero_std": 0.96875, "grad_norm": 0.4880352589785924, "kl": 0.1168212890625, "learning_rate": 4.987926867107095e-06, "loss": 0.0045, "num_tokens": 18036748.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.9375, "completions/max_length": 1841.0, "completions/max_terminated_length": 1841.0, "completions/mean_length": 492.521484375, "completions/mean_terminated_length": 484.14959716796875, "completions/min_length": 145.0, "completions/min_terminated_length": 145.0, "epoch": 0.029550033579583614, "frac_reward_zero_std": 0.96875, "grad_norm": 5.555072095753837, "kl": 0.425537109375, "learning_rate": 4.986372137660078e-06, "loss": 0.0105, "num_tokens": 18605991.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.921875, "completions/max_length": 1231.0, "completions/max_terminated_length": 1231.0, "completions/mean_length": 470.818359375, "completions/mean_terminated_length": 466.25048828125, "completions/min_length": 140.0, "completions/min_terminated_length": 140.0, "epoch": 0.030445489142601298, "frac_reward_zero_std": 1.0, "grad_norm": 0.061669261382861086, "kl": 0.1256103515625, "learning_rate": 4.984723574517165e-06, "loss": 0.0013, "num_tokens": 19125818.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.890625, "completions/max_length": 1387.0, "completions/max_terminated_length": 1387.0, "completions/mean_length": 532.9296875, "completions/mean_terminated_length": 525.10498046875, "completions/min_length": 172.0, "completions/min_terminated_length": 172.0, "epoch": 0.03134094470561898, "frac_reward_zero_std": 0.90625, "grad_norm": 17.57107852556159, "kl": 0.1639404296875, "learning_rate": 4.9829812468487655e-06, "loss": 0.0174, "num_tokens": 19718790.0, "reward": 0.09941406548023224, "reward_std": 0.0023437500931322575, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.994140625, "rewards/format_reward/std": 0.07639661431312561, "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": -6.734375, "completions/max_length": 1184.0, "completions/max_terminated_length": 1184.0, "completions/mean_length": 516.59765625, "completions/mean_terminated_length": 496.25860595703125, "completions/min_length": 188.0, "completions/min_terminated_length": 188.0, "epoch": 0.03223640026863667, "frac_reward_zero_std": 0.9375, "grad_norm": 17.78488865540282, "kl": 0.3154296875, "learning_rate": 4.981145227759457e-06, "loss": 0.0204, "num_tokens": 20301240.0, "reward": 0.09726562350988388, "reward_std": 0.0020409778226166964, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.97265625, "rewards/format_reward/std": 0.16324250400066376, "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": -6.84375, "completions/max_length": 1759.0, "completions/max_terminated_length": 1759.0, "completions/mean_length": 550.166015625, "completions/mean_terminated_length": 536.2271118164062, "completions/min_length": 180.0, "completions/min_terminated_length": 180.0, "epoch": 0.03313185583165435, "frac_reward_zero_std": 0.90625, "grad_norm": 7806.320146811803, "kl": 572.0, "learning_rate": 4.979215594284924e-06, "loss": 5.7231, "num_tokens": 20888077.0, "reward": 0.09941406548023224, "reward_std": 0.0023437500931322575, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.994140625, "rewards/format_reward/std": 0.07639661431312561, "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": -6.890625, "completions/max_length": 2048.0, "completions/max_terminated_length": 1769.0, "completions/mean_length": 570.6875, "completions/mean_terminated_length": 558.4415893554688, "completions/min_length": 100.0, "completions/min_terminated_length": 100.0, "epoch": 0.03402731139467204, "frac_reward_zero_std": 0.9375, "grad_norm": 29.0123039860226, "kl": 2.9072265625, "learning_rate": 4.977192427388722e-06, "loss": 0.051, "num_tokens": 21505389.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "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": -6.90625, "completions/max_length": 1847.0, "completions/max_terminated_length": 1847.0, "completions/mean_length": 538.337890625, "completions/mean_terminated_length": 531.9466552734375, "completions/min_length": 113.0, "completions/min_terminated_length": 113.0, "epoch": 0.03492276695768973, "frac_reward_zero_std": 0.90625, "grad_norm": 1.830223160421892, "kl": 0.1162109375, "learning_rate": 4.9750758119588824e-06, "loss": -0.0029, "num_tokens": 22082218.0, "reward": 0.09941406548023224, "reward_std": 0.0023437500931322575, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.994140625, "rewards/format_reward/std": 0.07639661431312561, "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": -6.84375, "completions/max_length": 1363.0, "completions/max_terminated_length": 1363.0, "completions/mean_length": 560.26953125, "completions/mean_terminated_length": 548.7888793945312, "completions/min_length": 186.0, "completions/min_terminated_length": 186.0, "epoch": 0.03581822252070741, "frac_reward_zero_std": 0.96875, "grad_norm": 5.189694636533056, "kl": 0.1761474609375, "learning_rate": 4.972865836804349e-06, "loss": 0.006, "num_tokens": 22659380.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.9375, "completions/max_length": 1279.0, "completions/max_terminated_length": 1279.0, "completions/mean_length": 478.83203125, "completions/mean_terminated_length": 475.21063232421875, "completions/min_length": 129.0, "completions/min_terminated_length": 129.0, "epoch": 0.036713678083725096, "frac_reward_zero_std": 0.96875, "grad_norm": 0.08930606384058017, "kl": 0.115966796875, "learning_rate": 4.970562594651254e-06, "loss": -0.0013, "num_tokens": 23177790.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.921875, "completions/max_length": 1525.0, "completions/max_terminated_length": 1525.0, "completions/mean_length": 544.818359375, "completions/mean_terminated_length": 535.7534790039062, "completions/min_length": 139.0, "completions/min_terminated_length": 139.0, "epoch": 0.03760913364674278, "frac_reward_zero_std": 0.96875, "grad_norm": 0.12171028839751209, "kl": 0.1319580078125, "learning_rate": 4.968166182139026e-06, "loss": 0.0031, "num_tokens": 23784705.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.796875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1370.0, "completions/mean_length": 490.720703125, "completions/mean_terminated_length": 470.34869384765625, "completions/min_length": 85.0, "completions/min_terminated_length": 85.0, "epoch": 0.038504589209760465, "frac_reward_zero_std": 0.96875, "grad_norm": 7.7636935355164045, "kl": 0.246826171875, "learning_rate": 4.9656766998163306e-06, "loss": 0.0094, "num_tokens": 24299730.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.921875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1637.0, "completions/mean_length": 527.525390625, "completions/mean_terminated_length": 520.1400756835938, "completions/min_length": 150.0, "completions/min_terminated_length": 150.0, "epoch": 0.03940004477277815, "frac_reward_zero_std": 0.96875, "grad_norm": 4.974293908259754, "kl": 0.6517333984375, "learning_rate": 4.963094252136865e-06, "loss": 0.0236, "num_tokens": 24869215.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.890625, "completions/max_length": 2048.0, "completions/max_terminated_length": 1675.0, "completions/mean_length": 517.080078125, "completions/mean_terminated_length": 506.8396301269531, "completions/min_length": 143.0, "completions/min_terminated_length": 143.0, "epoch": 0.040295500335795834, "frac_reward_zero_std": 0.90625, "grad_norm": 5.969358530129748, "kl": 0.1383056640625, "learning_rate": 4.960418947454958e-06, "loss": 0.0292, "num_tokens": 25461880.0, "reward": 0.09941406548023224, "reward_std": 0.0023437500931322575, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.994140625, "rewards/format_reward/std": 0.07639661431312561, "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": -6.9375, "completions/max_length": 1767.0, "completions/max_terminated_length": 1767.0, "completions/mean_length": 547.33203125, "completions/mean_terminated_length": 540.8327026367188, "completions/min_length": 75.0, "completions/min_terminated_length": 75.0, "epoch": 0.04119095589881352, "frac_reward_zero_std": 0.96875, "grad_norm": 3.964515594688607, "kl": 0.09521484375, "learning_rate": 4.957650898021038e-06, "loss": 0.0095, "num_tokens": 26040930.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.875, "completions/max_length": 2048.0, "completions/max_terminated_length": 2035.0, "completions/mean_length": 569.171875, "completions/mean_terminated_length": 545.6984252929688, "completions/min_length": 55.0, "completions/min_terminated_length": 55.0, "epoch": 0.04208641146183121, "frac_reward_zero_std": 0.90625, "grad_norm": 3.633363104209369, "kl": 0.6241455078125, "learning_rate": 4.954790219976915e-06, "loss": 0.0428, "num_tokens": 26644090.0, "reward": 0.09921875596046448, "reward_std": 0.0026298905722796917, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.9921875, "rewards/format_reward/std": 0.08812850713729858, "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": -6.671875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1566.0, "completions/mean_length": 581.9375, "completions/mean_terminated_length": 558.0386962890625, "completions/min_length": 129.0, "completions/min_terminated_length": 129.0, "epoch": 0.042981867024848894, "frac_reward_zero_std": 0.90625, "grad_norm": 1.3607360038408776, "kl": 0.390625, "learning_rate": 4.95183703335091e-06, "loss": 0.0081, "num_tokens": 27297130.0, "reward": 0.0966796875, "reward_std": 0.0023437500931322575, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.966796875, "rewards/format_reward/std": 0.17934183776378632, "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": -6.9375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1945.0, "completions/mean_length": 546.5859375, "completions/mean_terminated_length": 539.001953125, "completions/min_length": 128.0, "completions/min_terminated_length": 128.0, "epoch": 0.04387732258786658, "frac_reward_zero_std": 1.0, "grad_norm": 0.9432769696523784, "kl": 0.1865234375, "learning_rate": 4.948791462052819e-06, "loss": 0.0019, "num_tokens": 27855910.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.9375, "completions/max_length": 1071.0, "completions/max_terminated_length": 1071.0, "completions/mean_length": 510.458984375, "completions/mean_terminated_length": 506.47833251953125, "completions/min_length": 160.0, "completions/min_terminated_length": 160.0, "epoch": 0.04477277815088426, "frac_reward_zero_std": 0.96875, "grad_norm": 0.08737982579415644, "kl": 0.091796875, "learning_rate": 4.945653633868716e-06, "loss": 0.0053, "num_tokens": 28437265.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.953125, "completions/max_length": 1840.0, "completions/max_terminated_length": 1840.0, "completions/mean_length": 559.35546875, "completions/mean_terminated_length": 553.75048828125, "completions/min_length": 115.0, "completions/min_terminated_length": 115.0, "epoch": 0.04566823371390195, "frac_reward_zero_std": 0.96875, "grad_norm": 0.06582492357236211, "kl": 0.0966796875, "learning_rate": 4.942423680455584e-06, "loss": 0.0007, "num_tokens": 29032375.0, "reward": 0.099609375, "reward_std": 0.0010673906654119492, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "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": -6.859375, "completions/max_length": 1732.0, "completions/max_terminated_length": 1732.0, "completions/mean_length": 529.138671875, "completions/mean_terminated_length": 513.461181640625, "completions/min_length": 137.0, "completions/min_terminated_length": 137.0, "epoch": 0.04656368927691963, "frac_reward_zero_std": 0.96875, "grad_norm": 0.24454712354121277, "kl": 0.193115234375, "learning_rate": 4.939101737335802e-06, "loss": 0.012, "num_tokens": 29592686.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.9375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1830.0, "completions/mean_length": 514.66796875, "completions/mean_terminated_length": 507.16534423828125, "completions/min_length": 125.0, "completions/min_terminated_length": 125.0, "epoch": 0.047459144839937316, "frac_reward_zero_std": 0.96875, "grad_norm": 0.07209160390847483, "kl": 0.0982666015625, "learning_rate": 4.935687943891447e-06, "loss": 0.0171, "num_tokens": 30153252.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.90625, "completions/max_length": 1958.0, "completions/max_terminated_length": 1958.0, "completions/mean_length": 556.134765625, "completions/mean_terminated_length": 545.2826538085938, "completions/min_length": 74.0, "completions/min_terminated_length": 74.0, "epoch": 0.048354600402955, "frac_reward_zero_std": 0.96875, "grad_norm": 1.2182649822769658, "kl": 0.1866455078125, "learning_rate": 4.932182443358458e-06, "loss": 0.013, "num_tokens": 30773801.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.921875, "completions/max_length": 1496.0, "completions/max_terminated_length": 1496.0, "completions/mean_length": 520.10546875, "completions/mean_terminated_length": 513.6646728515625, "completions/min_length": 160.0, "completions/min_terminated_length": 160.0, "epoch": 0.04925005596597269, "frac_reward_zero_std": 1.0, "grad_norm": 0.06929038708329119, "kl": 0.1387939453125, "learning_rate": 4.928585382820616e-06, "loss": 0.0014, "num_tokens": 31299967.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.96875, "completions/max_length": 1904.0, "completions/max_terminated_length": 1904.0, "completions/mean_length": 542.359375, "completions/mean_terminated_length": 540.060791015625, "completions/min_length": 185.0, "completions/min_terminated_length": 185.0, "epoch": 0.050145511528990376, "frac_reward_zero_std": 0.96875, "grad_norm": 0.15563688859048186, "kl": 0.104736328125, "learning_rate": 4.924896913203376e-06, "loss": -0.0022, "num_tokens": 31888535.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.984375, "completions/max_length": 1506.0, "completions/max_terminated_length": 1506.0, "completions/mean_length": 516.908203125, "completions/mean_terminated_length": 515.5968627929688, "completions/min_length": 153.0, "completions/min_terminated_length": 153.0, "epoch": 0.05104096709200806, "frac_reward_zero_std": 0.9375, "grad_norm": 0.09565916384806829, "kl": 0.07861328125, "learning_rate": 4.921117189267535e-06, "loss": 0.0044, "num_tokens": 32461608.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "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": -6.96875, "completions/max_length": 1916.0, "completions/max_terminated_length": 1916.0, "completions/mean_length": 525.482421875, "completions/mean_terminated_length": 521.7431640625, "completions/min_length": 130.0, "completions/min_terminated_length": 130.0, "epoch": 0.051936422655025745, "frac_reward_zero_std": 0.96875, "grad_norm": 0.0748699769594785, "kl": 0.1156005859375, "learning_rate": 4.917246369602742e-06, "loss": 0.0002, "num_tokens": 33044847.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.984375, "completions/max_length": 2001.0, "completions/max_terminated_length": 2001.0, "completions/mean_length": 514.23046875, "completions/mean_terminated_length": 513.3111572265625, "completions/min_length": 188.0, "completions/min_terminated_length": 188.0, "epoch": 0.05283187821804343, "frac_reward_zero_std": 0.96875, "grad_norm": 0.03709928067002307, "kl": 0.079833984375, "learning_rate": 4.9132846166208355e-06, "loss": -0.0006, "num_tokens": 33620501.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1458.0, "completions/mean_length": 511.310546875, "completions/mean_terminated_length": 505.2843322753906, "completions/min_length": 118.0, "completions/min_terminated_length": 118.0, "epoch": 0.053727333781061114, "frac_reward_zero_std": 0.875, "grad_norm": 0.13840494930716188, "kl": 0.06787109375, "learning_rate": 4.9092320965490365e-06, "loss": 0.0211, "num_tokens": 34206116.0, "reward": 0.09902343899011612, "reward_std": 0.003411140525713563, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.990234375, "rewards/format_reward/std": 0.09843364357948303, "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": -7.0, "completions/max_length": 1895.0, "completions/max_terminated_length": 1895.0, "completions/mean_length": 518.853515625, "completions/mean_terminated_length": 518.853515625, "completions/min_length": 121.0, "completions/min_terminated_length": 121.0, "epoch": 0.0546227893440788, "frac_reward_zero_std": 0.875, "grad_norm": 0.14872642942212774, "kl": 0.0692138671875, "learning_rate": 4.905088979422971e-06, "loss": 0.0128, "num_tokens": 34793785.0, "reward": 0.09921875596046448, "reward_std": 0.0031250000465661287, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.9921875, "rewards/format_reward/std": 0.08812850713729858, "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": -6.984375, "completions/max_length": 1878.0, "completions/max_terminated_length": 1878.0, "completions/mean_length": 474.60546875, "completions/mean_terminated_length": 473.1917724609375, "completions/min_length": 145.0, "completions/min_terminated_length": 145.0, "epoch": 0.05551824490709648, "frac_reward_zero_std": 0.90625, "grad_norm": 0.12988434784735217, "kl": 0.0841064453125, "learning_rate": 4.900855439079536e-06, "loss": -0.0012, "num_tokens": 35339935.0, "reward": 0.09941406548023224, "reward_std": 0.0023437500931322575, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.994140625, "rewards/format_reward/std": 0.07639661431312561, "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": -7.0, "completions/max_length": 1579.0, "completions/max_terminated_length": 1579.0, "completions/mean_length": 504.658203125, "completions/mean_terminated_length": 504.658203125, "completions/min_length": 140.0, "completions/min_terminated_length": 140.0, "epoch": 0.056413700470114174, "frac_reward_zero_std": 0.9375, "grad_norm": 0.07899678142978885, "kl": 0.06787109375, "learning_rate": 4.8965316531496055e-06, "loss": -0.0002, "num_tokens": 35911392.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1658.0, "completions/mean_length": 541.5859375, "completions/mean_terminated_length": 535.678466796875, "completions/min_length": 141.0, "completions/min_terminated_length": 141.0, "epoch": 0.05730915603313186, "frac_reward_zero_std": 0.84375, "grad_norm": 0.1638638882495904, "kl": 0.06365966796875, "learning_rate": 4.892117803050578e-06, "loss": 0.0252, "num_tokens": 36471548.0, "reward": 0.09882812947034836, "reward_std": 0.004192390479147434, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.98828125, "rewards/format_reward/std": 0.10772226005792618, "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1190.0, "completions/mean_length": 509.482421875, "completions/mean_terminated_length": 505.13140869140625, "completions/min_length": 116.0, "completions/min_terminated_length": 116.0, "epoch": 0.05820461159614954, "frac_reward_zero_std": 0.90625, "grad_norm": 0.12491007191362062, "kl": 0.0806884765625, "learning_rate": 4.887614073978761e-06, "loss": 0.0189, "num_tokens": 37031171.0, "reward": 0.09921875596046448, "reward_std": 0.0026298905722796917, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.9921875, "rewards/format_reward/std": 0.08812850713729858, "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": -7.0, "completions/max_length": 1841.0, "completions/max_terminated_length": 1841.0, "completions/mean_length": 521.2890625, "completions/mean_terminated_length": 521.2890625, "completions/min_length": 98.0, "completions/min_terminated_length": 98.0, "epoch": 0.05910006715916723, "frac_reward_zero_std": 0.96875, "grad_norm": 0.08381886765190119, "kl": 0.06219482421875, "learning_rate": 4.883020654901609e-06, "loss": 0.0019, "num_tokens": 37627703.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -7.0, "completions/max_length": 1381.0, "completions/max_terminated_length": 1381.0, "completions/mean_length": 477.830078125, "completions/mean_terminated_length": 477.830078125, "completions/min_length": 157.0, "completions/min_terminated_length": 157.0, "epoch": 0.05999552272218491, "frac_reward_zero_std": 1.0, "grad_norm": 0.013399015519228851, "kl": 0.0623779296875, "learning_rate": 4.878337738549785e-06, "loss": 0.0006, "num_tokens": 38178656.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.984375, "completions/max_length": 1351.0, "completions/max_terminated_length": 1351.0, "completions/mean_length": 559.1171875, "completions/mean_terminated_length": 558.129150390625, "completions/min_length": 203.0, "completions/min_terminated_length": 203.0, "epoch": 0.060890978285202596, "frac_reward_zero_std": 1.0, "grad_norm": 0.02053915419301268, "kl": 0.0797119140625, "learning_rate": 4.873565521409082e-06, "loss": 0.0008, "num_tokens": 38780044.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -7.0, "completions/max_length": 1736.0, "completions/max_terminated_length": 1736.0, "completions/mean_length": 528.73046875, "completions/mean_terminated_length": 528.73046875, "completions/min_length": 131.0, "completions/min_terminated_length": 131.0, "epoch": 0.06178643384822028, "frac_reward_zero_std": 1.0, "grad_norm": 0.013023668854051146, "kl": 0.06231689453125, "learning_rate": 4.868704203712173e-06, "loss": 0.0006, "num_tokens": 39342786.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.9375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1704.0, "completions/mean_length": 539.8671875, "completions/mean_terminated_length": 529.748046875, "completions/min_length": 180.0, "completions/min_terminated_length": 180.0, "epoch": 0.06268188941123796, "frac_reward_zero_std": 0.875, "grad_norm": 7.439993648948339, "kl": 0.08404541015625, "learning_rate": 4.86375398943021e-06, "loss": 0.0415, "num_tokens": 39922430.0, "reward": 0.09921875596046448, "reward_std": 0.0031250000465661287, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.9921875, "rewards/format_reward/std": 0.08812850713729858, "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": -7.0, "completions/max_length": 1456.0, "completions/max_terminated_length": 1456.0, "completions/mean_length": 529.509765625, "completions/mean_terminated_length": 529.509765625, "completions/min_length": 195.0, "completions/min_terminated_length": 195.0, "epoch": 0.06357734497425566, "frac_reward_zero_std": 1.0, "grad_norm": 0.012082899535339748, "kl": 0.06158447265625, "learning_rate": 4.858715086264274e-06, "loss": 0.0006, "num_tokens": 40501891.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -7.0, "completions/max_length": 1235.0, "completions/max_terminated_length": 1235.0, "completions/mean_length": 563.224609375, "completions/mean_terminated_length": 563.224609375, "completions/min_length": 210.0, "completions/min_terminated_length": 210.0, "epoch": 0.06447280053727333, "frac_reward_zero_std": 0.90625, "grad_norm": 0.10284201346946772, "kl": 0.05999755859375, "learning_rate": 4.853587705636646e-06, "loss": 0.0043, "num_tokens": 41123062.0, "reward": 0.09941406548023224, "reward_std": 0.0023437500931322575, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.994140625, "rewards/format_reward/std": 0.07639661431312561, "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": -7.0, "completions/max_length": 1747.0, "completions/max_terminated_length": 1747.0, "completions/mean_length": 533.79296875, "completions/mean_terminated_length": 533.79296875, "completions/min_length": 195.0, "completions/min_terminated_length": 195.0, "epoch": 0.06536825610029103, "frac_reward_zero_std": 0.96875, "grad_norm": 0.050213646166851394, "kl": 0.0599365234375, "learning_rate": 4.84837206268195e-06, "loss": 0.0, "num_tokens": 41681452.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -7.0, "completions/max_length": 1085.0, "completions/max_terminated_length": 1085.0, "completions/mean_length": 509.712890625, "completions/mean_terminated_length": 509.712890625, "completions/min_length": 160.0, "completions/min_terminated_length": 160.0, "epoch": 0.0662637116633087, "frac_reward_zero_std": 0.96875, "grad_norm": 0.06633850089342387, "kl": 0.05963134765625, "learning_rate": 4.8430683762381195e-06, "loss": 0.0061, "num_tokens": 42219641.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -7.0, "completions/max_length": 1555.0, "completions/max_terminated_length": 1555.0, "completions/mean_length": 574.283203125, "completions/mean_terminated_length": 574.283203125, "completions/min_length": 200.0, "completions/min_terminated_length": 200.0, "epoch": 0.0671591672263264, "frac_reward_zero_std": 1.0, "grad_norm": 0.020666631584363913, "kl": 0.060791015625, "learning_rate": 4.837676868837213e-06, "loss": 0.0006, "num_tokens": 42824586.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -7.0, "completions/max_length": 1942.0, "completions/max_terminated_length": 1942.0, "completions/mean_length": 521.0546875, "completions/mean_terminated_length": 521.0546875, "completions/min_length": 139.0, "completions/min_terminated_length": 139.0, "epoch": 0.06805462278934409, "frac_reward_zero_std": 0.96875, "grad_norm": 0.06331685772788986, "kl": 0.05889892578125, "learning_rate": 4.832197766696085e-06, "loss": -0.0, "num_tokens": 43379718.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -7.0, "completions/max_length": 1598.0, "completions/max_terminated_length": 1598.0, "completions/mean_length": 568.814453125, "completions/mean_terminated_length": 568.814453125, "completions/min_length": 215.0, "completions/min_terminated_length": 215.0, "epoch": 0.06895007835236176, "frac_reward_zero_std": 0.9375, "grad_norm": 0.07366287417422632, "kl": 0.0562744140625, "learning_rate": 4.826631299706887e-06, "loss": -0.0016, "num_tokens": 43978151.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "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": -7.0, "completions/max_length": 1806.0, "completions/max_terminated_length": 1806.0, "completions/mean_length": 569.267578125, "completions/mean_terminated_length": 569.267578125, "completions/min_length": 220.0, "completions/min_terminated_length": 220.0, "epoch": 0.06984553391537945, "frac_reward_zero_std": 0.90625, "grad_norm": 0.11848160599061058, "kl": 0.058349609375, "learning_rate": 4.820977701427424e-06, "loss": 0.0152, "num_tokens": 44552064.0, "reward": 0.09941406548023224, "reward_std": 0.0023437500931322575, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.994140625, "rewards/format_reward/std": 0.07639661431312561, "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": -7.0, "completions/max_length": 1567.0, "completions/max_terminated_length": 1567.0, "completions/mean_length": 557.267578125, "completions/mean_terminated_length": 557.267578125, "completions/min_length": 181.0, "completions/min_terminated_length": 181.0, "epoch": 0.07074098947839713, "frac_reward_zero_std": 1.0, "grad_norm": 0.011448547949088674, "kl": 0.06097412109375, "learning_rate": 4.81523720907136e-06, "loss": 0.0006, "num_tokens": 45158809.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -7.0, "completions/max_length": 2009.0, "completions/max_terminated_length": 2009.0, "completions/mean_length": 552.33984375, "completions/mean_terminated_length": 552.33984375, "completions/min_length": 151.0, "completions/min_terminated_length": 151.0, "epoch": 0.07163644504141482, "frac_reward_zero_std": 0.96875, "grad_norm": 0.056084177201828476, "kl": 0.0584716796875, "learning_rate": 4.809410063498254e-06, "loss": -0.0002, "num_tokens": 45754263.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -7.0, "completions/max_length": 1401.0, "completions/max_terminated_length": 1401.0, "completions/mean_length": 543.6796875, "completions/mean_terminated_length": 543.6796875, "completions/min_length": 188.0, "completions/min_terminated_length": 188.0, "epoch": 0.0725319006044325, "frac_reward_zero_std": 0.9375, "grad_norm": 0.10338543772762039, "kl": 0.0606689453125, "learning_rate": 4.8034965092034656e-06, "loss": 0.0038, "num_tokens": 46356163.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "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": -6.984375, "completions/max_length": 1176.0, "completions/max_terminated_length": 1176.0, "completions/mean_length": 584.32421875, "completions/mean_terminated_length": 583.5518798828125, "completions/min_length": 236.0, "completions/min_terminated_length": 236.0, "epoch": 0.07342735616745019, "frac_reward_zero_std": 0.96875, "grad_norm": 0.0722417078888751, "kl": 0.0614013671875, "learning_rate": 4.797496794307889e-06, "loss": 0.0038, "num_tokens": 46960281.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1328.0, "completions/mean_length": 599.443359375, "completions/mean_terminated_length": 596.6085815429688, "completions/min_length": 209.0, "completions/min_terminated_length": 209.0, "epoch": 0.07432281173046787, "frac_reward_zero_std": 0.96875, "grad_norm": 0.04419533464180944, "kl": 0.0595703125, "learning_rate": 4.791411170547545e-06, "loss": 0.0122, "num_tokens": 47594764.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -7.0, "completions/max_length": 1370.0, "completions/max_terminated_length": 1370.0, "completions/mean_length": 567.396484375, "completions/mean_terminated_length": 567.396484375, "completions/min_length": 202.0, "completions/min_terminated_length": 202.0, "epoch": 0.07521826729348556, "frac_reward_zero_std": 1.0, "grad_norm": 0.008048576448064306, "kl": 0.0582275390625, "learning_rate": 4.785239893263017e-06, "loss": 0.0006, "num_tokens": 48214007.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -7.0, "completions/max_length": 1268.0, "completions/max_terminated_length": 1268.0, "completions/mean_length": 568.2421875, "completions/mean_terminated_length": 568.2421875, "completions/min_length": 188.0, "completions/min_terminated_length": 188.0, "epoch": 0.07611372285650325, "frac_reward_zero_std": 0.90625, "grad_norm": 0.13977703925477217, "kl": 0.06085205078125, "learning_rate": 4.778983221388742e-06, "loss": 0.0049, "num_tokens": 48800115.0, "reward": 0.09941406548023224, "reward_std": 0.0023437500931322575, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.994140625, "rewards/format_reward/std": 0.07639661431312561, "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1454.0, "completions/mean_length": 582.78515625, "completions/mean_terminated_length": 579.9177856445312, "completions/min_length": 172.0, "completions/min_terminated_length": 172.0, "epoch": 0.07700917841952093, "frac_reward_zero_std": 0.96875, "grad_norm": 0.07934026261985073, "kl": 0.0589599609375, "learning_rate": 4.77264141744214e-06, "loss": 0.0097, "num_tokens": 49419445.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -7.0, "completions/max_length": 1647.0, "completions/max_terminated_length": 1647.0, "completions/mean_length": 590.314453125, "completions/mean_terminated_length": 590.314453125, "completions/min_length": 230.0, "completions/min_terminated_length": 230.0, "epoch": 0.07790463398253862, "frac_reward_zero_std": 0.96875, "grad_norm": 0.06933720029794796, "kl": 0.0592041015625, "learning_rate": 4.766214747512603e-06, "loss": 0.001, "num_tokens": 50031558.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.984375, "completions/max_length": 1328.0, "completions/max_terminated_length": 1328.0, "completions/mean_length": 559.21875, "completions/mean_terminated_length": 558.4931640625, "completions/min_length": 208.0, "completions/min_terminated_length": 208.0, "epoch": 0.0788000895455563, "frac_reward_zero_std": 1.0, "grad_norm": 0.15169183218007004, "kl": 0.07476806640625, "learning_rate": 4.759703481250331e-06, "loss": 0.0007, "num_tokens": 50636678.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -7.0, "completions/max_length": 1333.0, "completions/max_terminated_length": 1333.0, "completions/mean_length": 577.947265625, "completions/mean_terminated_length": 577.947265625, "completions/min_length": 210.0, "completions/min_terminated_length": 210.0, "epoch": 0.07969554510857399, "frac_reward_zero_std": 1.0, "grad_norm": 0.007228029815201077, "kl": 0.058837890625, "learning_rate": 4.753107891855015e-06, "loss": 0.0006, "num_tokens": 51286507.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -7.0, "completions/max_length": 1744.0, "completions/max_terminated_length": 1744.0, "completions/mean_length": 552.662109375, "completions/mean_terminated_length": 552.662109375, "completions/min_length": 150.0, "completions/min_terminated_length": 150.0, "epoch": 0.08059100067159167, "frac_reward_zero_std": 0.96875, "grad_norm": 0.05217256333372253, "kl": 0.06201171875, "learning_rate": 4.746428256064375e-06, "loss": -0.001, "num_tokens": 51888542.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -7.0, "completions/max_length": 1383.0, "completions/max_terminated_length": 1383.0, "completions/mean_length": 529.599609375, "completions/mean_terminated_length": 529.599609375, "completions/min_length": 136.0, "completions/min_terminated_length": 136.0, "epoch": 0.08148645623460936, "frac_reward_zero_std": 0.96875, "grad_norm": 0.07897485826909563, "kl": 0.0584716796875, "learning_rate": 4.7396648541425534e-06, "loss": 0.0055, "num_tokens": 52460673.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -7.0, "completions/max_length": 1405.0, "completions/max_terminated_length": 1405.0, "completions/mean_length": 534.078125, "completions/mean_terminated_length": 534.078125, "completions/min_length": 159.0, "completions/min_terminated_length": 159.0, "epoch": 0.08238191179762704, "frac_reward_zero_std": 1.0, "grad_norm": 0.009069004256641576, "kl": 0.058837890625, "learning_rate": 4.732817969868348e-06, "loss": 0.0006, "num_tokens": 53053129.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -7.0, "completions/max_length": 1509.0, "completions/max_terminated_length": 1509.0, "completions/mean_length": 581.345703125, "completions/mean_terminated_length": 581.345703125, "completions/min_length": 175.0, "completions/min_terminated_length": 175.0, "epoch": 0.08327736736064473, "frac_reward_zero_std": 1.0, "grad_norm": 0.008041339436883252, "kl": 0.05926513671875, "learning_rate": 4.7258878905233095e-06, "loss": 0.0006, "num_tokens": 53676490.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -7.0, "completions/max_length": 1690.0, "completions/max_terminated_length": 1690.0, "completions/mean_length": 570.1796875, "completions/mean_terminated_length": 570.1796875, "completions/min_length": 181.0, "completions/min_terminated_length": 181.0, "epoch": 0.08417282292366242, "frac_reward_zero_std": 0.96875, "grad_norm": 0.08004068477244176, "kl": 0.06011962890625, "learning_rate": 4.718874906879688e-06, "loss": 0.0048, "num_tokens": 54286198.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -7.0, "completions/max_length": 1286.0, "completions/max_terminated_length": 1286.0, "completions/mean_length": 503.283203125, "completions/mean_terminated_length": 503.283203125, "completions/min_length": 156.0, "completions/min_terminated_length": 156.0, "epoch": 0.0850682784866801, "frac_reward_zero_std": 1.0, "grad_norm": 0.007110972110569579, "kl": 0.05828857421875, "learning_rate": 4.711779313188231e-06, "loss": 0.0006, "num_tokens": 54808951.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -7.0, "completions/max_length": 1328.0, "completions/max_terminated_length": 1328.0, "completions/mean_length": 575.7421875, "completions/mean_terminated_length": 575.7421875, "completions/min_length": 194.0, "completions/min_terminated_length": 194.0, "epoch": 0.08596373404969779, "frac_reward_zero_std": 0.96875, "grad_norm": 0.06146080952581339, "kl": 0.058349609375, "learning_rate": 4.70460140716584e-06, "loss": 0.0038, "num_tokens": 55425011.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -7.0, "completions/max_length": 1760.0, "completions/max_terminated_length": 1760.0, "completions/mean_length": 510.607421875, "completions/mean_terminated_length": 510.607421875, "completions/min_length": 141.0, "completions/min_terminated_length": 141.0, "epoch": 0.08685918961271547, "frac_reward_zero_std": 1.0, "grad_norm": 0.013320757255101928, "kl": 0.05645751953125, "learning_rate": 4.697341489983076e-06, "loss": 0.0006, "num_tokens": 55965402.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -7.0, "completions/max_length": 1890.0, "completions/max_terminated_length": 1890.0, "completions/mean_length": 565.958984375, "completions/mean_terminated_length": 565.958984375, "completions/min_length": 188.0, "completions/min_terminated_length": 188.0, "epoch": 0.08775464517573316, "frac_reward_zero_std": 1.0, "grad_norm": 0.007573822836148033, "kl": 0.05853271484375, "learning_rate": 4.6899998662515215e-06, "loss": 0.0006, "num_tokens": 56571765.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -7.0, "completions/max_length": 1135.0, "completions/max_terminated_length": 1135.0, "completions/mean_length": 503.03125, "completions/mean_terminated_length": 503.03125, "completions/min_length": 175.0, "completions/min_terminated_length": 175.0, "epoch": 0.08865010073875083, "frac_reward_zero_std": 1.0, "grad_norm": 0.007334410202363631, "kl": 0.05633544921875, "learning_rate": 4.682576844011007e-06, "loss": 0.0006, "num_tokens": 57124213.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -7.0, "completions/max_length": 1658.0, "completions/max_terminated_length": 1658.0, "completions/mean_length": 536.2421875, "completions/mean_terminated_length": 536.2421875, "completions/min_length": 134.0, "completions/min_terminated_length": 134.0, "epoch": 0.08954555630176853, "frac_reward_zero_std": 1.0, "grad_norm": 0.007211066238880028, "kl": 0.05853271484375, "learning_rate": 4.675072734716678e-06, "loss": 0.0006, "num_tokens": 57682705.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -7.0, "completions/max_length": 1632.0, "completions/max_terminated_length": 1632.0, "completions/mean_length": 554.22265625, "completions/mean_terminated_length": 554.22265625, "completions/min_length": 199.0, "completions/min_terminated_length": 199.0, "epoch": 0.09044101186478622, "frac_reward_zero_std": 1.0, "grad_norm": 0.006708160382694399, "kl": 0.05792236328125, "learning_rate": 4.667487853225931e-06, "loss": 0.0006, "num_tokens": 58310755.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1931.0, "completions/mean_length": 566.173828125, "completions/mean_terminated_length": 563.2739868164062, "completions/min_length": 143.0, "completions/min_terminated_length": 143.0, "epoch": 0.0913364674278039, "frac_reward_zero_std": 0.96875, "grad_norm": 0.07622570043853691, "kl": 0.05584716796875, "learning_rate": 4.659822517785203e-06, "loss": 0.0137, "num_tokens": 58888860.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -7.0, "completions/max_length": 1372.0, "completions/max_terminated_length": 1372.0, "completions/mean_length": 553.12109375, "completions/mean_terminated_length": 553.12109375, "completions/min_length": 212.0, "completions/min_terminated_length": 212.0, "epoch": 0.09223192299082159, "frac_reward_zero_std": 0.96875, "grad_norm": 0.049130696414956106, "kl": 0.05828857421875, "learning_rate": 4.6520770500166165e-06, "loss": -0.0001, "num_tokens": 59485690.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.984375, "completions/max_length": 1689.0, "completions/max_terminated_length": 1689.0, "completions/mean_length": 592.68359375, "completions/mean_terminated_length": 591.5635986328125, "completions/min_length": 156.0, "completions/min_terminated_length": 156.0, "epoch": 0.09312737855383926, "frac_reward_zero_std": 0.9375, "grad_norm": 4.817209315659024, "kl": 0.07928466796875, "learning_rate": 4.644251774904487e-06, "loss": 0.0076, "num_tokens": 60135864.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "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": -7.0, "completions/max_length": 1234.0, "completions/max_terminated_length": 1234.0, "completions/mean_length": 563.57421875, "completions/mean_terminated_length": 563.57421875, "completions/min_length": 168.0, "completions/min_terminated_length": 168.0, "epoch": 0.09402283411685695, "frac_reward_zero_std": 1.0, "grad_norm": 0.009712812135983481, "kl": 0.05499267578125, "learning_rate": 4.636347020781684e-06, "loss": 0.0006, "num_tokens": 60712014.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -7.0, "completions/max_length": 1801.0, "completions/max_terminated_length": 1801.0, "completions/mean_length": 551.984375, "completions/mean_terminated_length": 551.984375, "completions/min_length": 161.0, "completions/min_terminated_length": 161.0, "epoch": 0.09491828967987463, "frac_reward_zero_std": 0.96875, "grad_norm": 0.06067404264171204, "kl": 0.05560302734375, "learning_rate": 4.6283631193158605e-06, "loss": 0.0012, "num_tokens": 61273174.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.984375, "completions/max_length": 1733.0, "completions/max_terminated_length": 1733.0, "completions/mean_length": 565.373046875, "completions/mean_terminated_length": 564.3267822265625, "completions/min_length": 179.0, "completions/min_terminated_length": 179.0, "epoch": 0.09581374524289232, "frac_reward_zero_std": 1.0, "grad_norm": 19.400452290618507, "kl": 3.197265625, "learning_rate": 4.620300405495532e-06, "loss": 0.032, "num_tokens": 61852677.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1473.0, "completions/mean_length": 554.125, "completions/mean_terminated_length": 551.2015380859375, "completions/min_length": 159.0, "completions/min_terminated_length": 159.0, "epoch": 0.09670920080591, "frac_reward_zero_std": 0.96875, "grad_norm": 0.0524692551884981, "kl": 0.05401611328125, "learning_rate": 4.612159217616022e-06, "loss": 0.0146, "num_tokens": 62462309.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -7.0, "completions/max_length": 1408.0, "completions/max_terminated_length": 1408.0, "completions/mean_length": 546.9453125, "completions/mean_terminated_length": 546.9453125, "completions/min_length": 240.0, "completions/min_terminated_length": 240.0, "epoch": 0.09760465636892769, "frac_reward_zero_std": 1.0, "grad_norm": 0.006616016440406459, "kl": 0.05621337890625, "learning_rate": 4.603939897265268e-06, "loss": 0.0006, "num_tokens": 63050249.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -7.0, "completions/max_length": 1746.0, "completions/max_terminated_length": 1746.0, "completions/mean_length": 540.046875, "completions/mean_terminated_length": 540.046875, "completions/min_length": 175.0, "completions/min_terminated_length": 175.0, "epoch": 0.09850011193194538, "frac_reward_zero_std": 1.0, "grad_norm": 0.006058315501498899, "kl": 0.0535888671875, "learning_rate": 4.595642789309492e-06, "loss": 0.0005, "num_tokens": 63619009.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -7.0, "completions/max_length": 1615.0, "completions/max_terminated_length": 1615.0, "completions/mean_length": 546.98046875, "completions/mean_terminated_length": 546.98046875, "completions/min_length": 154.0, "completions/min_terminated_length": 154.0, "epoch": 0.09939556749496306, "frac_reward_zero_std": 1.0, "grad_norm": 0.006228639344980519, "kl": 0.05523681640625, "learning_rate": 4.587268241878724e-06, "loss": 0.0006, "num_tokens": 64165991.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.984375, "completions/max_length": 1374.0, "completions/max_terminated_length": 1374.0, "completions/mean_length": 555.0703125, "completions/mean_terminated_length": 554.5205688476562, "completions/min_length": 236.0, "completions/min_terminated_length": 236.0, "epoch": 0.10029102305798075, "frac_reward_zero_std": 0.96875, "grad_norm": 1.2290379791712172, "kl": 0.05560302734375, "learning_rate": 4.578816606352205e-06, "loss": 0.0045, "num_tokens": 64768443.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.984375, "completions/max_length": 1258.0, "completions/max_terminated_length": 1258.0, "completions/mean_length": 563.681640625, "completions/mean_terminated_length": 562.8062744140625, "completions/min_length": 189.0, "completions/min_terminated_length": 189.0, "epoch": 0.10118647862099843, "frac_reward_zero_std": 0.96875, "grad_norm": 16.255998645519178, "kl": 0.74365234375, "learning_rate": 4.570288237343632e-06, "loss": 0.0048, "num_tokens": 65357352.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -7.0, "completions/max_length": 1490.0, "completions/max_terminated_length": 1490.0, "completions/mean_length": 586.935546875, "completions/mean_terminated_length": 586.935546875, "completions/min_length": 191.0, "completions/min_terminated_length": 191.0, "epoch": 0.10208193418401612, "frac_reward_zero_std": 0.96875, "grad_norm": 0.0774700444571494, "kl": 0.0572509765625, "learning_rate": 4.561683492686289e-06, "loss": 0.0038, "num_tokens": 65968535.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -7.0, "completions/max_length": 1247.0, "completions/max_terminated_length": 1247.0, "completions/mean_length": 541.302734375, "completions/mean_terminated_length": 541.302734375, "completions/min_length": 184.0, "completions/min_terminated_length": 184.0, "epoch": 0.1029773897470338, "frac_reward_zero_std": 1.0, "grad_norm": 0.00852963260021217, "kl": 0.0596923828125, "learning_rate": 4.5530027334180285e-06, "loss": 0.0006, "num_tokens": 66535154.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -7.0, "completions/max_length": 1153.0, "completions/max_terminated_length": 1153.0, "completions/mean_length": 504.041015625, "completions/mean_terminated_length": 504.041015625, "completions/min_length": 142.0, "completions/min_terminated_length": 142.0, "epoch": 0.10387284531005149, "frac_reward_zero_std": 1.0, "grad_norm": 0.00898169383540532, "kl": 0.060791015625, "learning_rate": 4.544246323766122e-06, "loss": 0.0006, "num_tokens": 67104631.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -7.0, "completions/max_length": 1247.0, "completions/max_terminated_length": 1247.0, "completions/mean_length": 537.59375, "completions/mean_terminated_length": 537.59375, "completions/min_length": 176.0, "completions/min_terminated_length": 176.0, "epoch": 0.10476830087306917, "frac_reward_zero_std": 1.0, "grad_norm": 0.009282105832995153, "kl": 0.06085205078125, "learning_rate": 4.535414631131983e-06, "loss": 0.0006, "num_tokens": 67668119.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -7.0, "completions/max_length": 1405.0, "completions/max_terminated_length": 1405.0, "completions/mean_length": 516.166015625, "completions/mean_terminated_length": 516.166015625, "completions/min_length": 130.0, "completions/min_terminated_length": 130.0, "epoch": 0.10566375643608686, "frac_reward_zero_std": 0.9375, "grad_norm": 0.07276575023389194, "kl": 0.06048583984375, "learning_rate": 4.526508026075746e-06, "loss": -0.0023, "num_tokens": 68238044.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "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": -7.0, "completions/max_length": 1659.0, "completions/max_terminated_length": 1659.0, "completions/mean_length": 546.119140625, "completions/mean_terminated_length": 546.119140625, "completions/min_length": 126.0, "completions/min_terminated_length": 126.0, "epoch": 0.10655921199910455, "frac_reward_zero_std": 1.0, "grad_norm": 0.007709260903345854, "kl": 0.056884765625, "learning_rate": 4.517526882300721e-06, "loss": 0.0006, "num_tokens": 68832137.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -7.0, "completions/max_length": 1518.0, "completions/max_terminated_length": 1518.0, "completions/mean_length": 510.16796875, "completions/mean_terminated_length": 510.16796875, "completions/min_length": 142.0, "completions/min_terminated_length": 142.0, "epoch": 0.10745466756212223, "frac_reward_zero_std": 0.9375, "grad_norm": 0.12374567038474613, "kl": 0.05853271484375, "learning_rate": 4.508471576637713e-06, "loss": 0.0052, "num_tokens": 69416831.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "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": -7.0, "completions/max_length": 1343.0, "completions/max_terminated_length": 1343.0, "completions/mean_length": 517.318359375, "completions/mean_terminated_length": 517.318359375, "completions/min_length": 192.0, "completions/min_terminated_length": 192.0, "epoch": 0.10835012312513992, "frac_reward_zero_std": 1.0, "grad_norm": 0.007982824080905008, "kl": 0.0570068359375, "learning_rate": 4.499342489029211e-06, "loss": 0.0006, "num_tokens": 69983394.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -7.0, "completions/max_length": 1176.0, "completions/max_terminated_length": 1176.0, "completions/mean_length": 503.2109375, "completions/mean_terminated_length": 503.2109375, "completions/min_length": 103.0, "completions/min_terminated_length": 103.0, "epoch": 0.1092455786881576, "frac_reward_zero_std": 0.96875, "grad_norm": 0.05137313735665502, "kl": 0.0572509765625, "learning_rate": 4.490140002513449e-06, "loss": -0.0022, "num_tokens": 70554302.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.984375, "completions/max_length": 1103.0, "completions/max_terminated_length": 1103.0, "completions/mean_length": 492.255859375, "completions/mean_terminated_length": 491.2739562988281, "completions/min_length": 90.0, "completions/min_terminated_length": 90.0, "epoch": 0.11014103425117529, "frac_reward_zero_std": 0.96875, "grad_norm": 6.6575044592016654, "kl": 0.06549072265625, "learning_rate": 4.48086450320833e-06, "loss": 0.0053, "num_tokens": 71102657.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -7.0, "completions/max_length": 1139.0, "completions/max_terminated_length": 1139.0, "completions/mean_length": 501.88671875, "completions/mean_terminated_length": 501.88671875, "completions/min_length": 196.0, "completions/min_terminated_length": 196.0, "epoch": 0.11103648981419297, "frac_reward_zero_std": 0.9375, "grad_norm": 0.13448036932392413, "kl": 0.05462646484375, "learning_rate": 4.4715163802952266e-06, "loss": 0.0022, "num_tokens": 71662455.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "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": -7.0, "completions/max_length": 1883.0, "completions/max_terminated_length": 1883.0, "completions/mean_length": 558.08203125, "completions/mean_terminated_length": 558.08203125, "completions/min_length": 210.0, "completions/min_terminated_length": 210.0, "epoch": 0.11193194537721066, "frac_reward_zero_std": 1.0, "grad_norm": 0.008863906624678173, "kl": 0.05322265625, "learning_rate": 4.462096026002655e-06, "loss": 0.0005, "num_tokens": 72236449.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -7.0, "completions/max_length": 1646.0, "completions/max_terminated_length": 1646.0, "completions/mean_length": 523.720703125, "completions/mean_terminated_length": 523.720703125, "completions/min_length": 183.0, "completions/min_terminated_length": 183.0, "epoch": 0.11282740094022835, "frac_reward_zero_std": 1.0, "grad_norm": 0.009601045986280915, "kl": 0.0523681640625, "learning_rate": 4.4526038355898144e-06, "loss": 0.0005, "num_tokens": 72817346.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -7.0, "completions/max_length": 1414.0, "completions/max_terminated_length": 1414.0, "completions/mean_length": 511.84375, "completions/mean_terminated_length": 511.84375, "completions/min_length": 188.0, "completions/min_terminated_length": 188.0, "epoch": 0.11372285650324603, "frac_reward_zero_std": 1.0, "grad_norm": 0.01150702874361092, "kl": 0.05126953125, "learning_rate": 4.4430402073300035e-06, "loss": 0.0005, "num_tokens": 73390642.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -7.0, "completions/max_length": 1298.0, "completions/max_terminated_length": 1298.0, "completions/mean_length": 535.732421875, "completions/mean_terminated_length": 535.732421875, "completions/min_length": 228.0, "completions/min_terminated_length": 228.0, "epoch": 0.11461831206626372, "frac_reward_zero_std": 1.0, "grad_norm": 0.9117448141973743, "kl": 0.12908935546875, "learning_rate": 4.433405542493909e-06, "loss": 0.0013, "num_tokens": 73959529.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.9375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1485.0, "completions/mean_length": 606.228515625, "completions/mean_terminated_length": 594.8759765625, "completions/min_length": 275.0, "completions/min_terminated_length": 275.0, "epoch": 0.1155137676292814, "frac_reward_zero_std": 1.0, "grad_norm": 0.014323382678470751, "kl": 0.0703125, "learning_rate": 4.4237002453327734e-06, "loss": 0.0007, "num_tokens": 74576526.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.828125, "completions/max_length": 2048.0, "completions/max_terminated_length": 1993.0, "completions/mean_length": 693.314453125, "completions/mean_terminated_length": 663.5708618164062, "completions/min_length": 261.0, "completions/min_terminated_length": 261.0, "epoch": 0.11640922319229909, "frac_reward_zero_std": 0.8125, "grad_norm": 0.18814500549038302, "kl": 0.106689453125, "learning_rate": 4.4139247230614245e-06, "loss": 0.0036, "num_tokens": 75240063.0, "reward": 0.09882812201976776, "reward_std": 0.004687500186264515, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.98828125, "rewards/format_reward/std": 0.10772226005792618, "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": -6.53125, "completions/max_length": 2048.0, "completions/max_terminated_length": 2036.0, "completions/mean_length": 822.12109375, "completions/mean_terminated_length": 745.8215942382812, "completions/min_length": 282.0, "completions/min_terminated_length": 282.0, "epoch": 0.11730467875531676, "frac_reward_zero_std": 1.0, "grad_norm": 0.02343943908233584, "kl": 0.1158447265625, "learning_rate": 4.404079385841201e-06, "loss": 0.0012, "num_tokens": 75948637.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.015625, "completions/max_length": 2048.0, "completions/max_terminated_length": 2039.0, "completions/mean_length": 994.90625, "completions/mean_terminated_length": 847.144775390625, "completions/min_length": 234.0, "completions/min_terminated_length": 234.0, "epoch": 0.11820013431833445, "frac_reward_zero_std": 0.90625, "grad_norm": 0.07306244420161156, "kl": 0.11376953125, "learning_rate": 4.394164646762734e-06, "loss": 0.0054, "num_tokens": 76751421.0, "reward": 0.09941406548023224, "reward_std": 0.0023437500931322575, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.994140625, "rewards/format_reward/std": 0.07639661431312561, "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": -5.84375, "completions/max_length": 2048.0, "completions/max_terminated_length": 2009.0, "completions/mean_length": 1017.630859375, "completions/mean_terminated_length": 843.5501708984375, "completions/min_length": 270.0, "completions/min_terminated_length": 270.0, "epoch": 0.11909558988135213, "frac_reward_zero_std": 0.96875, "grad_norm": 0.04774455505526359, "kl": 0.11669921875, "learning_rate": 4.384180921828618e-06, "loss": -0.0014, "num_tokens": 77631184.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -5.1875, "completions/max_length": 2048.0, "completions/max_terminated_length": 2044.0, "completions/mean_length": 1152.490234375, "completions/mean_terminated_length": 890.169189453125, "completions/min_length": 266.0, "completions/min_terminated_length": 266.0, "epoch": 0.11999104544436982, "frac_reward_zero_std": 0.96875, "grad_norm": 0.07160446510060933, "kl": 0.120361328125, "learning_rate": 4.374128629935955e-06, "loss": 0.0006, "num_tokens": 78525467.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -4.59375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1966.0, "completions/mean_length": 1246.322265625, "completions/mean_terminated_length": 901.4664306640625, "completions/min_length": 203.0, "completions/min_terminated_length": 203.0, "epoch": 0.12088650100738751, "frac_reward_zero_std": 0.9375, "grad_norm": 0.05630071606380346, "kl": 0.11083984375, "learning_rate": 4.364008192858781e-06, "loss": -0.003, "num_tokens": 79475744.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "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": -4.5625, "completions/max_length": 2048.0, "completions/max_terminated_length": 2027.0, "completions/mean_length": 1288.12890625, "completions/mean_terminated_length": 955.1516723632812, "completions/min_length": 243.0, "completions/min_terminated_length": 243.0, "epoch": 0.12178195657040519, "frac_reward_zero_std": 0.875, "grad_norm": 795332.7231700029, "kl": 14656.082153320312, "learning_rate": 4.353820035230366e-06, "loss": 145.7659, "num_tokens": 80433074.0, "reward": 0.09921875596046448, "reward_std": 0.0031250000465661287, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.9921875, "rewards/format_reward/std": 0.08812850713729858, "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": -4.265625, "completions/max_length": 2048.0, "completions/max_terminated_length": 2030.0, "completions/mean_length": 1294.751953125, "completions/mean_terminated_length": 903.5994262695312, "completions/min_length": 226.0, "completions/min_terminated_length": 226.0, "epoch": 0.12267741213342288, "frac_reward_zero_std": 0.90625, "grad_norm": 0.1687795917678753, "kl": 0.126708984375, "learning_rate": 4.3435645845254e-06, "loss": 0.0049, "num_tokens": 81393939.0, "reward": 0.09941406548023224, "reward_std": 0.0023437500931322575, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.994140625, "rewards/format_reward/std": 0.07639661431312561, "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": -4.015625, "completions/max_length": 2048.0, "completions/max_terminated_length": 2025.0, "completions/mean_length": 1392.58203125, "completions/mean_terminated_length": 1002.5980834960938, "completions/min_length": 301.0, "completions/min_terminated_length": 301.0, "epoch": 0.12357286769644056, "frac_reward_zero_std": 0.78125, "grad_norm": 0.09891611716752852, "kl": 0.1055908203125, "learning_rate": 4.333242271042054e-06, "loss": -0.0155, "num_tokens": 82458253.0, "reward": 0.09843750298023224, "reward_std": 0.005754890851676464, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.984375, "rewards/format_reward/std": 0.12414088100194931, "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": -4.3125, "completions/max_length": 2048.0, "completions/max_terminated_length": 2020.0, "completions/mean_length": 1325.6328125, "completions/mean_terminated_length": 960.2000122070312, "completions/min_length": 202.0, "completions/min_terminated_length": 202.0, "epoch": 0.12446832325945825, "frac_reward_zero_std": 0.75, "grad_norm": 0.11212612719912256, "kl": 0.1063232421875, "learning_rate": 4.32285352788393e-06, "loss": -0.005, "num_tokens": 83456625.0, "reward": 0.09824219346046448, "reward_std": 0.006536140572279692, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.982421875, "rewards/format_reward/std": 0.13154059648513794, "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": -4.15625, "completions/max_length": 2048.0, "completions/max_terminated_length": 2022.0, "completions/mean_length": 1346.60546875, "completions/mean_terminated_length": 959.7757568359375, "completions/min_length": 272.0, "completions/min_terminated_length": 272.0, "epoch": 0.12536377882247593, "frac_reward_zero_std": 0.78125, "grad_norm": 12694.24016060589, "kl": 618.6192626953125, "learning_rate": 4.312398790941882e-06, "loss": 6.1707, "num_tokens": 84461159.0, "reward": 0.09843750298023224, "reward_std": 0.005754890851676464, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.984375, "rewards/format_reward/std": 0.12414088100194931, "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": -4.671875, "completions/max_length": 2048.0, "completions/max_terminated_length": 2011.0, "completions/mean_length": 1241.689453125, "completions/mean_terminated_length": 910.7245483398438, "completions/min_length": 230.0, "completions/min_terminated_length": 230.0, "epoch": 0.1262592343854936, "frac_reward_zero_std": 0.9375, "grad_norm": 0.059238726705022914, "kl": 0.1033935546875, "learning_rate": 4.301878498875735e-06, "loss": 0.0002, "num_tokens": 85383528.0, "reward": 0.09941406548023224, "reward_std": 0.0018486406188458204, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.994140625, "rewards/format_reward/std": 0.07639661431312561, "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": -4.421875, "completions/max_length": 2048.0, "completions/max_terminated_length": 2011.0, "completions/mean_length": 1294.9375, "completions/mean_terminated_length": 936.85302734375, "completions/min_length": 306.0, "completions/min_terminated_length": 306.0, "epoch": 0.1271546899485113, "frac_reward_zero_std": 0.84375, "grad_norm": 0.08422896350695738, "kl": 0.1038818359375, "learning_rate": 4.291293093095873e-06, "loss": -0.0012, "num_tokens": 86330200.0, "reward": 0.09882812947034836, "reward_std": 0.004192390479147434, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.98828125, "rewards/format_reward/std": 0.10772226005792618, "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": -4.203125, "completions/max_length": 2048.0, "completions/max_terminated_length": 2007.0, "completions/mean_length": 1359.435546875, "completions/mean_terminated_length": 989.3063354492188, "completions/min_length": 276.0, "completions/min_terminated_length": 276.0, "epoch": 0.128050145511529, "frac_reward_zero_std": 0.84375, "grad_norm": 1.0100953665705108, "kl": 0.123291015625, "learning_rate": 4.280643017744723e-06, "loss": -0.0056, "num_tokens": 87369495.0, "reward": 0.09902344644069672, "reward_std": 0.00390625, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.990234375, "rewards/format_reward/std": 0.09843364357948303, "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": -3.859375, "completions/max_length": 2048.0, "completions/max_terminated_length": 2021.0, "completions/mean_length": 1424.6328125, "completions/mean_terminated_length": 1021.7492065429688, "completions/min_length": 348.0, "completions/min_terminated_length": 348.0, "epoch": 0.12894560107454667, "frac_reward_zero_std": 0.6875, "grad_norm": 0.09935539337282194, "kl": 0.0885009765625, "learning_rate": 4.269928719678117e-06, "loss": 0.0017, "num_tokens": 88405371.0, "reward": 0.09765625, "reward_std": 0.008384780958294868, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.9765625, "rewards/format_reward/std": 0.15143637359142303, "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": -4.140625, "completions/max_length": 2048.0, "completions/max_terminated_length": 2035.0, "completions/mean_length": 1348.701171875, "completions/mean_terminated_length": 959.7294921875, "completions/min_length": 338.0, "completions/min_terminated_length": 338.0, "epoch": 0.12984105663756437, "frac_reward_zero_std": 0.9375, "grad_norm": 0.2595208434649135, "kl": 0.113037109375, "learning_rate": 4.2591506484465426e-06, "loss": 0.0059, "num_tokens": 89399538.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "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": -4.203125, "completions/max_length": 2048.0, "completions/max_terminated_length": 2045.0, "completions/mean_length": 1366.169921875, "completions/mean_terminated_length": 999.6607055664062, "completions/min_length": 247.0, "completions/min_terminated_length": 247.0, "epoch": 0.13073651220058205, "frac_reward_zero_std": 0.875, "grad_norm": 0.08052216200621251, "kl": 0.1043701171875, "learning_rate": 4.248309256276283e-06, "loss": 0.006, "num_tokens": 90401529.0, "reward": 0.09921875596046448, "reward_std": 0.0031250000465661287, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.9921875, "rewards/format_reward/std": 0.08812850713729858, "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": -4.359375, "completions/max_length": 2048.0, "completions/max_terminated_length": 2045.0, "completions/mean_length": 1320.82421875, "completions/mean_terminated_length": 962.5364990234375, "completions/min_length": 255.0, "completions/min_terminated_length": 255.0, "epoch": 0.13163196776359973, "frac_reward_zero_std": 0.78125, "grad_norm": 2.41982629512539, "kl": 0.103759765625, "learning_rate": 4.23740499805044e-06, "loss": 0.0062, "num_tokens": 91347151.0, "reward": 0.0986328125, "reward_std": 0.00546875037252903, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.986328125, "rewards/format_reward/std": 0.1162383034825325, "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": -4.515625, "completions/max_length": 2048.0, "completions/max_terminated_length": 2046.0, "completions/mean_length": 1281.87109375, "completions/mean_terminated_length": 936.7875366210938, "completions/min_length": 227.0, "completions/min_terminated_length": 227.0, "epoch": 0.1325274233266174, "frac_reward_zero_std": 0.6875, "grad_norm": 1.0229642621220172, "kl": 0.1658935546875, "learning_rate": 4.22643833128985e-06, "loss": 0.0131, "num_tokens": 92310541.0, "reward": 0.09687500447034836, "reward_std": 0.009193411096930504, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.96875, "rewards/format_reward/std": 0.17416280508041382, "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": -4.21875, "completions/max_length": 2048.0, "completions/max_terminated_length": 2035.0, "completions/mean_length": 1364.658203125, "completions/mean_terminated_length": 1000.4820556640625, "completions/min_length": 315.0, "completions/min_terminated_length": 315.0, "epoch": 0.1334228788896351, "frac_reward_zero_std": 0.75, "grad_norm": 0.11093213327135282, "kl": 0.0858154296875, "learning_rate": 4.215409716133885e-06, "loss": -0.0018, "num_tokens": 93312318.0, "reward": 0.09824219346046448, "reward_std": 0.006536140572279692, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.982421875, "rewards/format_reward/std": 0.13154059648513794, "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": -4.5625, "completions/max_length": 2048.0, "completions/max_terminated_length": 2046.0, "completions/mean_length": 1338.37890625, "completions/mean_terminated_length": 1027.42138671875, "completions/min_length": 209.0, "completions/min_terminated_length": 209.0, "epoch": 0.1343183344526528, "frac_reward_zero_std": 0.8125, "grad_norm": 0.3995251198579641, "kl": 0.0902099609375, "learning_rate": 4.204319615321151e-06, "loss": 0.0086, "num_tokens": 94314160.0, "reward": 0.09882812947034836, "reward_std": 0.004687500186264515, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.98828125, "rewards/format_reward/std": 0.10772226005792618, "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": -4.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 2040.0, "completions/mean_length": 1252.2109375, "completions/mean_terminated_length": 984.1775512695312, "completions/min_length": 215.0, "completions/min_terminated_length": 215.0, "epoch": 0.13521379001567047, "frac_reward_zero_std": 0.75, "grad_norm": 0.11590978634626535, "kl": 0.08642578125, "learning_rate": 4.193168494170065e-06, "loss": -0.0085, "num_tokens": 95272492.0, "reward": 0.09804687649011612, "reward_std": 0.006822281051427126, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.98046875, "rewards/format_reward/std": 0.1385180652141571, "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": -5.046875, "completions/max_length": 2048.0, "completions/max_terminated_length": 2045.0, "completions/mean_length": 1240.875, "completions/mean_terminated_length": 980.1757202148438, "completions/min_length": 241.0, "completions/min_terminated_length": 241.0, "epoch": 0.13610924557868817, "frac_reward_zero_std": 0.90625, "grad_norm": 0.06993291127200373, "kl": 0.120849609375, "learning_rate": 4.181956820559339e-06, "loss": -0.0054, "num_tokens": 96201068.0, "reward": 0.09921875596046448, "reward_std": 0.0026298905722796917, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.9921875, "rewards/format_reward/std": 0.08812850713729858, "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": -5.578125, "completions/max_length": 2048.0, "completions/max_terminated_length": 2033.0, "completions/mean_length": 1163.3671875, "completions/mean_terminated_length": 972.1520385742188, "completions/min_length": 219.0, "completions/min_terminated_length": 219.0, "epoch": 0.13700470114170585, "frac_reward_zero_std": 0.84375, "grad_norm": 0.0861319462004464, "kl": 0.0950927734375, "learning_rate": 4.170685064908342e-06, "loss": 0.0035, "num_tokens": 97089800.0, "reward": 0.09902343899011612, "reward_std": 0.00390625, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.990234375, "rewards/format_reward/std": 0.09843364357948303, "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": -5.3125, "completions/max_length": 2048.0, "completions/max_terminated_length": 2035.0, "completions/mean_length": 1177.39453125, "completions/mean_terminated_length": 944.6583862304688, "completions/min_length": 238.0, "completions/min_terminated_length": 238.0, "epoch": 0.13790015670472353, "frac_reward_zero_std": 0.78125, "grad_norm": 0.26491805736565843, "kl": 0.1044921875, "learning_rate": 4.159353700157365e-06, "loss": -0.0087, "num_tokens": 97982770.0, "reward": 0.09843750298023224, "reward_std": 0.005754890851676464, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.984375, "rewards/format_reward/std": 0.12414088100194931, "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": -5.796875, "completions/max_length": 2048.0, "completions/max_terminated_length": 2033.0, "completions/mean_length": 1122.61328125, "completions/mean_terminated_length": 958.8092041015625, "completions/min_length": 274.0, "completions/min_terminated_length": 274.0, "epoch": 0.1387956122677412, "frac_reward_zero_std": 0.90625, "grad_norm": 0.0671134919909452, "kl": 0.0772705078125, "learning_rate": 4.14796320174778e-06, "loss": 0.0031, "num_tokens": 98844428.0, "reward": 0.09941406548023224, "reward_std": 0.0023437500931322575, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.994140625, "rewards/format_reward/std": 0.07639661431312561, "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": -5.6875, "completions/max_length": 2048.0, "completions/max_terminated_length": 2039.0, "completions/mean_length": 1108.546875, "completions/mean_terminated_length": 924.168212890625, "completions/min_length": 198.0, "completions/min_terminated_length": 198.0, "epoch": 0.1396910678307589, "frac_reward_zero_std": 0.8125, "grad_norm": 0.09175374952691938, "kl": 0.098876953125, "learning_rate": 4.136514047602087e-06, "loss": -0.0086, "num_tokens": 99727028.0, "reward": 0.09882812947034836, "reward_std": 0.004687500186264515, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.98828125, "rewards/format_reward/std": 0.10772226005792618, "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": -5.84375, "completions/max_length": 2048.0, "completions/max_terminated_length": 2020.0, "completions/mean_length": 1090.578125, "completions/mean_terminated_length": 928.8218994140625, "completions/min_length": 277.0, "completions/min_terminated_length": 277.0, "epoch": 0.14058652339377659, "frac_reward_zero_std": 0.9375, "grad_norm": 0.057742259396112945, "kl": 0.0733642578125, "learning_rate": 4.1250067181038635e-06, "loss": 0.0023, "num_tokens": 100603164.0, "reward": 0.09941406548023224, "reward_std": 0.0018486406188458204, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.994140625, "rewards/format_reward/std": 0.07639661431312561, "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": -5.640625, "completions/max_length": 2048.0, "completions/max_terminated_length": 2011.0, "completions/mean_length": 1118.96875, "completions/mean_terminated_length": 928.79052734375, "completions/min_length": 245.0, "completions/min_terminated_length": 245.0, "epoch": 0.14148197895679426, "frac_reward_zero_std": 1.0, "grad_norm": 0.009952836592016451, "kl": 0.0738525390625, "learning_rate": 4.113441696077608e-06, "loss": 0.0007, "num_tokens": 101509644.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.046875, "completions/max_length": 2048.0, "completions/max_terminated_length": 2047.0, "completions/mean_length": 1012.00390625, "completions/mean_terminated_length": 871.8802490234375, "completions/min_length": 216.0, "completions/min_terminated_length": 216.0, "epoch": 0.14237743451981194, "frac_reward_zero_std": 0.9375, "grad_norm": 0.14858531404881065, "kl": 0.076904296875, "learning_rate": 4.101819466768484e-06, "loss": -0.0044, "num_tokens": 102328510.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "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": -6.40625, "completions/max_length": 2048.0, "completions/max_terminated_length": 1991.0, "completions/mean_length": 921.74609375, "completions/mean_terminated_length": 831.4556274414062, "completions/min_length": 223.0, "completions/min_terminated_length": 223.0, "epoch": 0.14327289008282965, "frac_reward_zero_std": 0.75, "grad_norm": 0.1361238385560723, "kl": 0.11376953125, "learning_rate": 4.0901405178219535e-06, "loss": -0.0019, "num_tokens": 103101340.0, "reward": 0.09843750298023224, "reward_std": 0.0062500000931322575, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.984375, "rewards/format_reward/std": 0.12414088100194931, "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": -6.4375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1997.0, "completions/mean_length": 858.59765625, "completions/mean_terminated_length": 768.6428833007812, "completions/min_length": 174.0, "completions/min_terminated_length": 174.0, "epoch": 0.14416834564584732, "frac_reward_zero_std": 0.90625, "grad_norm": 0.056481508031488666, "kl": 0.07568359375, "learning_rate": 4.078405339263326e-06, "loss": -0.0145, "num_tokens": 103830654.0, "reward": 0.09941406548023224, "reward_std": 0.0023437500931322575, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.994140625, "rewards/format_reward/std": 0.07639661431312561, "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": -6.4375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1971.0, "completions/mean_length": 945.384765625, "completions/mean_terminated_length": 861.9937133789062, "completions/min_length": 234.0, "completions/min_terminated_length": 234.0, "epoch": 0.145063801208865, "frac_reward_zero_std": 0.9375, "grad_norm": 0.05837129013687458, "kl": 0.06982421875, "learning_rate": 4.06661442347719e-06, "loss": 0.0009, "num_tokens": 104648867.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "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": -6.734375, "completions/max_length": 2048.0, "completions/max_terminated_length": 2028.0, "completions/mean_length": 852.2265625, "completions/mean_terminated_length": 811.1596069335938, "completions/min_length": 195.0, "completions/min_terminated_length": 195.0, "epoch": 0.1459592567718827, "frac_reward_zero_std": 0.9375, "grad_norm": 0.079316699909495, "kl": 0.0712890625, "learning_rate": 4.054768265186758e-06, "loss": 0.0102, "num_tokens": 105394327.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "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": -6.6875, "completions/max_length": 2048.0, "completions/max_terminated_length": 2008.0, "completions/mean_length": 787.642578125, "completions/mean_terminated_length": 736.4085083007812, "completions/min_length": 228.0, "completions/min_terminated_length": 228.0, "epoch": 0.14685471233490038, "frac_reward_zero_std": 0.96875, "grad_norm": 0.034540728831484666, "kl": 0.085693359375, "learning_rate": 4.0428673614331036e-06, "loss": -0.0013, "num_tokens": 106085968.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.59375, "completions/max_length": 2048.0, "completions/max_terminated_length": 2048.0, "completions/mean_length": 830.478515625, "completions/mean_terminated_length": 765.3436279296875, "completions/min_length": 210.0, "completions/min_terminated_length": 210.0, "epoch": 0.14775016789791806, "frac_reward_zero_std": 0.90625, "grad_norm": 0.06310920256066563, "kl": 0.074951171875, "learning_rate": 4.030912211554316e-06, "loss": -0.0152, "num_tokens": 106846837.0, "reward": 0.09941406548023224, "reward_std": 0.0023437500931322575, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.994140625, "rewards/format_reward/std": 0.07639661431312561, "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": -6.640625, "completions/max_length": 2048.0, "completions/max_terminated_length": 1986.0, "completions/mean_length": 834.1953125, "completions/mean_terminated_length": 777.104248046875, "completions/min_length": 221.0, "completions/min_terminated_length": 221.0, "epoch": 0.14864562346093574, "frac_reward_zero_std": 1.0, "grad_norm": 0.011733653254552742, "kl": 0.0711669921875, "learning_rate": 4.018903317164539e-06, "loss": 0.0007, "num_tokens": 107568953.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.796875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1966.0, "completions/mean_length": 774.447265625, "completions/mean_terminated_length": 741.2685546875, "completions/min_length": 256.0, "completions/min_terminated_length": 256.0, "epoch": 0.14954107902395344, "frac_reward_zero_std": 0.96875, "grad_norm": 0.03409817940123074, "kl": 0.069580078125, "learning_rate": 4.006841182132932e-06, "loss": -0.0017, "num_tokens": 108281694.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.828125, "completions/max_length": 2048.0, "completions/max_terminated_length": 1971.0, "completions/mean_length": 756.509765625, "completions/mean_terminated_length": 728.1536865234375, "completions/min_length": 227.0, "completions/min_terminated_length": 227.0, "epoch": 0.15043653458697112, "frac_reward_zero_std": 0.96875, "grad_norm": 0.053464405396001835, "kl": 0.073486328125, "learning_rate": 3.9947263125625195e-06, "loss": -0.0007, "num_tokens": 108949523.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.671875, "completions/max_length": 2048.0, "completions/max_terminated_length": 2010.0, "completions/mean_length": 788.6875, "completions/mean_terminated_length": 734.826904296875, "completions/min_length": 208.0, "completions/min_terminated_length": 208.0, "epoch": 0.1513319901499888, "frac_reward_zero_std": 0.9375, "grad_norm": 0.07949072343589329, "kl": 0.0740966796875, "learning_rate": 3.982559216768967e-06, "loss": 0.0034, "num_tokens": 109670835.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "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": -6.796875, "completions/max_length": 2048.0, "completions/max_terminated_length": 2000.0, "completions/mean_length": 698.345703125, "completions/mean_terminated_length": 663.1843872070312, "completions/min_length": 149.0, "completions/min_terminated_length": 149.0, "epoch": 0.1522274457130065, "frac_reward_zero_std": 0.96875, "grad_norm": 0.045217268440116885, "kl": 0.0711669921875, "learning_rate": 3.970340405259245e-06, "loss": -0.0016, "num_tokens": 110321060.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.890625, "completions/max_length": 2048.0, "completions/max_terminated_length": 1813.0, "completions/mean_length": 644.583984375, "completions/mean_terminated_length": 625.1307373046875, "completions/min_length": 168.0, "completions/min_terminated_length": 168.0, "epoch": 0.15312290127602418, "frac_reward_zero_std": 0.96875, "grad_norm": 0.05776747187731656, "kl": 0.0753173828125, "learning_rate": 3.958070390710214e-06, "loss": 0.0006, "num_tokens": 110936255.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.9375, "completions/max_length": 2048.0, "completions/max_terminated_length": 2021.0, "completions/mean_length": 692.998046875, "completions/mean_terminated_length": 682.3287353515625, "completions/min_length": 236.0, "completions/min_terminated_length": 236.0, "epoch": 0.15401835683904186, "frac_reward_zero_std": 0.96875, "grad_norm": 0.056419783846101156, "kl": 0.072021484375, "learning_rate": 3.945749687947109e-06, "loss": 0.0005, "num_tokens": 111589950.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.921875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1706.0, "completions/mean_length": 645.02734375, "completions/mean_terminated_length": 631.1913452148438, "completions/min_length": 164.0, "completions/min_terminated_length": 164.0, "epoch": 0.15491381240205954, "frac_reward_zero_std": 1.0, "grad_norm": 0.008793484041353715, "kl": 0.0721435546875, "learning_rate": 3.933378813921942e-06, "loss": 0.0007, "num_tokens": 112188604.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.859375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1917.0, "completions/mean_length": 708.283203125, "completions/mean_terminated_length": 684.3120727539062, "completions/min_length": 149.0, "completions/min_terminated_length": 149.0, "epoch": 0.15580926796507724, "frac_reward_zero_std": 0.96875, "grad_norm": 0.041267501649098925, "kl": 0.0745849609375, "learning_rate": 3.920958287691811e-06, "loss": -0.0033, "num_tokens": 112857709.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.921875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1992.0, "completions/mean_length": 678.421875, "completions/mean_terminated_length": 664.9152221679688, "completions/min_length": 200.0, "completions/min_terminated_length": 200.0, "epoch": 0.15670472352809492, "frac_reward_zero_std": 0.9375, "grad_norm": 0.06445929796633103, "kl": 0.0740966796875, "learning_rate": 3.908488630397121e-06, "loss": -0.0031, "num_tokens": 113514933.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "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": -6.921875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1982.0, "completions/mean_length": 672.736328125, "completions/mean_terminated_length": 659.173583984375, "completions/min_length": 149.0, "completions/min_terminated_length": 149.0, "epoch": 0.1576001790911126, "frac_reward_zero_std": 0.96875, "grad_norm": 0.05116548843844329, "kl": 0.0716552734375, "learning_rate": 3.8959703652397175e-06, "loss": -0.0012, "num_tokens": 114166094.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.90625, "completions/max_length": 2048.0, "completions/max_terminated_length": 2028.0, "completions/mean_length": 710.96875, "completions/mean_terminated_length": 695.1541748046875, "completions/min_length": 146.0, "completions/min_terminated_length": 146.0, "epoch": 0.1584956346541303, "frac_reward_zero_std": 1.0, "grad_norm": 0.006938377171807984, "kl": 0.092529296875, "learning_rate": 3.883404017460935e-06, "loss": 0.0009, "num_tokens": 114829342.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1971.0, "completions/mean_length": 691.232421875, "completions/mean_terminated_length": 669.6964721679688, "completions/min_length": 251.0, "completions/min_terminated_length": 251.0, "epoch": 0.15939109021714798, "frac_reward_zero_std": 0.9375, "grad_norm": 0.0818158362296435, "kl": 0.0672607421875, "learning_rate": 3.870790114319559e-06, "loss": 0.0031, "num_tokens": 115473525.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "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": -6.890625, "completions/max_length": 2048.0, "completions/max_terminated_length": 1721.0, "completions/mean_length": 639.69140625, "completions/mean_terminated_length": 620.1702880859375, "completions/min_length": 156.0, "completions/min_terminated_length": 156.0, "epoch": 0.16028654578016566, "frac_reward_zero_std": 1.0, "grad_norm": 0.007890322312255954, "kl": 0.07080078125, "learning_rate": 3.858129185069701e-06, "loss": 0.0007, "num_tokens": 116083991.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.9375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1891.0, "completions/mean_length": 689.5546875, "completions/mean_terminated_length": 679.472412109375, "completions/min_length": 172.0, "completions/min_terminated_length": 172.0, "epoch": 0.16118200134318333, "frac_reward_zero_std": 0.9375, "grad_norm": 0.07882687764220778, "kl": 0.11083984375, "learning_rate": 3.845421760938597e-06, "loss": 0.0029, "num_tokens": 116761651.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "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": -6.953125, "completions/max_length": 2048.0, "completions/max_terminated_length": 2033.0, "completions/mean_length": 635.724609375, "completions/mean_terminated_length": 627.4008178710938, "completions/min_length": 233.0, "completions/min_terminated_length": 233.0, "epoch": 0.16207745690620104, "frac_reward_zero_std": 0.96875, "grad_norm": 0.0725677041475304, "kl": 0.070556640625, "learning_rate": 3.832668375104312e-06, "loss": -0.0004, "num_tokens": 117383574.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.921875, "completions/max_length": 2048.0, "completions/max_terminated_length": 2003.0, "completions/mean_length": 713.12109375, "completions/mean_terminated_length": 699.9566040039062, "completions/min_length": 205.0, "completions/min_terminated_length": 205.0, "epoch": 0.16297291246921872, "frac_reward_zero_std": 1.0, "grad_norm": 0.014271174770978515, "kl": 0.0697021484375, "learning_rate": 3.8198695626733725e-06, "loss": 0.0007, "num_tokens": 118063748.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.84375, "completions/max_length": 2048.0, "completions/max_terminated_length": 2040.0, "completions/mean_length": 713.150390625, "completions/mean_terminated_length": 686.5597534179688, "completions/min_length": 219.0, "completions/min_terminated_length": 219.0, "epoch": 0.1638683680322364, "frac_reward_zero_std": 0.90625, "grad_norm": 0.08507378122083202, "kl": 0.105224609375, "learning_rate": 3.8070258606583156e-06, "loss": -0.0037, "num_tokens": 118727089.0, "reward": 0.09941406548023224, "reward_std": 0.0023437500931322575, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.994140625, "rewards/format_reward/std": 0.07639661431312561, "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": -6.9375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1989.0, "completions/mean_length": 722.576171875, "completions/mean_terminated_length": 712.1397705078125, "completions/min_length": 184.0, "completions/min_terminated_length": 184.0, "epoch": 0.16476382359525407, "frac_reward_zero_std": 0.96875, "grad_norm": 0.07094156347361062, "kl": 0.0694580078125, "learning_rate": 3.7941378079551544e-06, "loss": 0.0022, "num_tokens": 119412936.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.90625, "completions/max_length": 2048.0, "completions/max_terminated_length": 1883.0, "completions/mean_length": 696.935546875, "completions/mean_terminated_length": 680.9150390625, "completions/min_length": 194.0, "completions/min_terminated_length": 194.0, "epoch": 0.16565927915827178, "frac_reward_zero_std": 1.0, "grad_norm": 0.007349496040761767, "kl": 0.0682373046875, "learning_rate": 3.7812059453207677e-06, "loss": 0.0007, "num_tokens": 120084967.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.9375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1895.0, "completions/mean_length": 676.48828125, "completions/mean_terminated_length": 665.68896484375, "completions/min_length": 196.0, "completions/min_terminated_length": 196.0, "epoch": 0.16655473472128945, "frac_reward_zero_std": 0.9375, "grad_norm": 0.07706499183432042, "kl": 0.067626953125, "learning_rate": 3.768230815350213e-06, "loss": -0.0007, "num_tokens": 120746113.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "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": -6.875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1893.0, "completions/mean_length": 636.73046875, "completions/mean_terminated_length": 616.3154907226562, "completions/min_length": 171.0, "completions/min_terminated_length": 171.0, "epoch": 0.16745019028430713, "frac_reward_zero_std": 0.96875, "grad_norm": 0.04911030768903344, "kl": 0.099853515625, "learning_rate": 3.7552129624539557e-06, "loss": -0.002, "num_tokens": 121372679.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.9375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1942.0, "completions/mean_length": 626.787109375, "completions/mean_terminated_length": 615.596435546875, "completions/min_length": 156.0, "completions/min_terminated_length": 156.0, "epoch": 0.16834564584732484, "frac_reward_zero_std": 0.9375, "grad_norm": 0.06941023633883373, "kl": 0.0716552734375, "learning_rate": 3.7421529328350316e-06, "loss": -0.0033, "num_tokens": 121996346.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "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": -6.921875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1995.0, "completions/mean_length": 690.220703125, "completions/mean_terminated_length": 676.8303833007812, "completions/min_length": 127.0, "completions/min_terminated_length": 127.0, "epoch": 0.16924110141034251, "frac_reward_zero_std": 0.9375, "grad_norm": 0.07568288588640923, "kl": 0.066162109375, "learning_rate": 3.7290512744661274e-06, "loss": 0.0084, "num_tokens": 122671259.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "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": -6.90625, "completions/max_length": 2048.0, "completions/max_terminated_length": 1983.0, "completions/mean_length": 697.283203125, "completions/mean_terminated_length": 681.266845703125, "completions/min_length": 197.0, "completions/min_terminated_length": 197.0, "epoch": 0.1701365569733602, "frac_reward_zero_std": 0.96875, "grad_norm": 0.049509900781209426, "kl": 0.0650634765625, "learning_rate": 3.715908537066589e-06, "loss": 0.001, "num_tokens": 123326652.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.9375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1783.0, "completions/mean_length": 626.283203125, "completions/mean_terminated_length": 615.0885620117188, "completions/min_length": 185.0, "completions/min_terminated_length": 185.0, "epoch": 0.17103201253637787, "frac_reward_zero_std": 1.0, "grad_norm": 0.007656796040386732, "kl": 0.066650390625, "learning_rate": 3.7027252720793538e-06, "loss": 0.0007, "num_tokens": 123951005.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.90625, "completions/max_length": 2048.0, "completions/max_terminated_length": 2035.0, "completions/mean_length": 665.228515625, "completions/mean_terminated_length": 648.83203125, "completions/min_length": 138.0, "completions/min_terminated_length": 138.0, "epoch": 0.17192746809939558, "frac_reward_zero_std": 0.96875, "grad_norm": 0.05792298135309963, "kl": 0.0677490234375, "learning_rate": 3.689502032647817e-06, "loss": -0.0015, "num_tokens": 124596706.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.9375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1688.0, "completions/mean_length": 620.7578125, "completions/mean_terminated_length": 609.5196533203125, "completions/min_length": 171.0, "completions/min_terminated_length": 171.0, "epoch": 0.17282292366241325, "frac_reward_zero_std": 1.0, "grad_norm": 0.007164808257495399, "kl": 0.06610107421875, "learning_rate": 3.6762393735926245e-06, "loss": 0.0007, "num_tokens": 125195014.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.9375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1790.0, "completions/mean_length": 638.5703125, "completions/mean_terminated_length": 627.472412109375, "completions/min_length": 230.0, "completions/min_terminated_length": 230.0, "epoch": 0.17371837922543093, "frac_reward_zero_std": 0.96875, "grad_norm": 0.060704628699271526, "kl": 0.06787109375, "learning_rate": 3.6629378513883852e-06, "loss": 0.0009, "num_tokens": 125829882.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1977.0, "completions/mean_length": 624.65625, "completions/mean_terminated_length": 619.0745239257812, "completions/min_length": 194.0, "completions/min_terminated_length": 194.0, "epoch": 0.17461383478844864, "frac_reward_zero_std": 1.0, "grad_norm": 0.007082540003861009, "kl": 0.066162109375, "learning_rate": 3.6495980241403307e-06, "loss": 0.0007, "num_tokens": 126427866.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 2033.0, "completions/mean_length": 701.626953125, "completions/mean_terminated_length": 698.9921875, "completions/min_length": 214.0, "completions/min_terminated_length": 214.0, "epoch": 0.1755092903514663, "frac_reward_zero_std": 1.0, "grad_norm": 0.006717488479532777, "kl": 0.0650634765625, "learning_rate": 3.636220451560896e-06, "loss": 0.0007, "num_tokens": 127094715.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.9375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1944.0, "completions/mean_length": 662.830078125, "completions/mean_terminated_length": 651.9232177734375, "completions/min_length": 205.0, "completions/min_terminated_length": 205.0, "epoch": 0.176404745914484, "frac_reward_zero_std": 0.9375, "grad_norm": 0.08069331728310686, "kl": 0.0665283203125, "learning_rate": 3.622805694946235e-06, "loss": -0.0005, "num_tokens": 127758052.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1890.0, "completions/mean_length": 676.98828125, "completions/mean_terminated_length": 671.61181640625, "completions/min_length": 250.0, "completions/min_terminated_length": 250.0, "epoch": 0.17730020147750167, "frac_reward_zero_std": 0.9375, "grad_norm": 0.07236099664779054, "kl": 0.068359375, "learning_rate": 3.609354317152667e-06, "loss": -0.0019, "num_tokens": 128406366.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1925.0, "completions/mean_length": 668.119140625, "completions/mean_terminated_length": 662.7078857421875, "completions/min_length": 179.0, "completions/min_terminated_length": 179.0, "epoch": 0.17819565704051937, "frac_reward_zero_std": 0.96875, "grad_norm": 0.040604973329934714, "kl": 0.0650634765625, "learning_rate": 3.595866882573063e-06, "loss": -0.0003, "num_tokens": 129045771.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.890625, "completions/max_length": 2048.0, "completions/max_terminated_length": 1728.0, "completions/mean_length": 622.59765625, "completions/mean_terminated_length": 602.839599609375, "completions/min_length": 122.0, "completions/min_terminated_length": 122.0, "epoch": 0.17909111260353705, "frac_reward_zero_std": 0.96875, "grad_norm": 0.03790785409368292, "kl": 0.078369140625, "learning_rate": 3.5823439571131675e-06, "loss": -0.0017, "num_tokens": 129675389.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1954.0, "completions/mean_length": 647.595703125, "completions/mean_terminated_length": 642.1039428710938, "completions/min_length": 184.0, "completions/min_terminated_length": 184.0, "epoch": 0.17998656816655473, "frac_reward_zero_std": 1.0, "grad_norm": 0.009018080934348364, "kl": 0.067138671875, "learning_rate": 3.5687861081678477e-06, "loss": 0.0007, "num_tokens": 130304526.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.953125, "completions/max_length": 2048.0, "completions/max_terminated_length": 1711.0, "completions/mean_length": 616.826171875, "completions/mean_terminated_length": 608.3909912109375, "completions/min_length": 212.0, "completions/min_terminated_length": 212.0, "epoch": 0.18088202372957243, "frac_reward_zero_std": 0.96875, "grad_norm": 0.04772213814585424, "kl": 0.068115234375, "learning_rate": 3.555193904597291e-06, "loss": -0.0004, "num_tokens": 130928037.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1809.0, "completions/mean_length": 642.734375, "completions/mean_terminated_length": 620.4285888671875, "completions/min_length": 158.0, "completions/min_terminated_length": 158.0, "epoch": 0.1817774792925901, "frac_reward_zero_std": 1.0, "grad_norm": 0.0082751790561264, "kl": 0.065185546875, "learning_rate": 3.541567916703138e-06, "loss": 0.0007, "num_tokens": 131538845.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.921875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1929.0, "completions/mean_length": 691.69921875, "completions/mean_terminated_length": 678.323486328125, "completions/min_length": 159.0, "completions/min_terminated_length": 159.0, "epoch": 0.1826729348556078, "frac_reward_zero_std": 0.96875, "grad_norm": 0.06259185264923466, "kl": 0.06427001953125, "learning_rate": 3.5279087162045517e-06, "loss": 0.009, "num_tokens": 132206467.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.953125, "completions/max_length": 2048.0, "completions/max_terminated_length": 2009.0, "completions/mean_length": 672.505859375, "completions/mean_terminated_length": 664.3988647460938, "completions/min_length": 245.0, "completions/min_terminated_length": 245.0, "epoch": 0.18356839041862547, "frac_reward_zero_std": 0.96875, "grad_norm": 0.07151737136940692, "kl": 0.0640869140625, "learning_rate": 3.5142168762142265e-06, "loss": 0.0023, "num_tokens": 132888022.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.859375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1765.0, "completions/mean_length": 700.201171875, "completions/mean_terminated_length": 676.08544921875, "completions/min_length": 193.0, "completions/min_terminated_length": 193.0, "epoch": 0.18446384598164317, "frac_reward_zero_std": 0.9375, "grad_norm": 0.08013199443660099, "kl": 0.067626953125, "learning_rate": 3.500492971214347e-06, "loss": 0.0001, "num_tokens": 133558285.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "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": -7.0, "completions/max_length": 1677.0, "completions/max_terminated_length": 1677.0, "completions/mean_length": 591.958984375, "completions/mean_terminated_length": 591.958984375, "completions/min_length": 200.0, "completions/min_terminated_length": 200.0, "epoch": 0.18535930154466085, "frac_reward_zero_std": 0.9375, "grad_norm": 0.07996600764674648, "kl": 0.06640625, "learning_rate": 3.48673757703248e-06, "loss": -0.0019, "num_tokens": 134162152.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1960.0, "completions/mean_length": 624.052734375, "completions/mean_terminated_length": 618.4686889648438, "completions/min_length": 250.0, "completions/min_terminated_length": 250.0, "epoch": 0.18625475710767853, "frac_reward_zero_std": 1.0, "grad_norm": 0.010916793033113123, "kl": 0.065185546875, "learning_rate": 3.472951270817418e-06, "loss": 0.0007, "num_tokens": 134795923.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.9375, "completions/max_length": 2048.0, "completions/max_terminated_length": 2023.0, "completions/mean_length": 617.498046875, "completions/mean_terminated_length": 606.2342529296875, "completions/min_length": 188.0, "completions/min_terminated_length": 188.0, "epoch": 0.1871502126706962, "frac_reward_zero_std": 0.96875, "grad_norm": 0.05456360479658491, "kl": 0.0648193359375, "learning_rate": 3.4591346310149578e-06, "loss": 0.0019, "num_tokens": 135414450.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.953125, "completions/max_length": 2048.0, "completions/max_terminated_length": 2038.0, "completions/mean_length": 662.5078125, "completions/mean_terminated_length": 654.3418579101562, "completions/min_length": 145.0, "completions/min_terminated_length": 145.0, "epoch": 0.1880456682337139, "frac_reward_zero_std": 0.90625, "grad_norm": 0.10789941880075726, "kl": 0.06365966796875, "learning_rate": 3.445288237343632e-06, "loss": 0.0034, "num_tokens": 136086934.0, "reward": 0.09941406548023224, "reward_std": 0.0023437500931322575, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.994140625, "rewards/format_reward/std": 0.07639661431312561, "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": -6.953125, "completions/max_length": 2048.0, "completions/max_terminated_length": 2008.0, "completions/mean_length": 717.3984375, "completions/mean_terminated_length": 709.5560302734375, "completions/min_length": 265.0, "completions/min_terminated_length": 265.0, "epoch": 0.18894112379673159, "frac_reward_zero_std": 1.0, "grad_norm": 0.0062856581992966785, "kl": 0.06195068359375, "learning_rate": 3.4314126707703895e-06, "loss": 0.0006, "num_tokens": 136787346.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 2039.0, "completions/mean_length": 625.490234375, "completions/mean_terminated_length": 619.9118041992188, "completions/min_length": 164.0, "completions/min_terminated_length": 164.0, "epoch": 0.18983657935974926, "frac_reward_zero_std": 0.96875, "grad_norm": 0.04953062525787495, "kl": 0.0645751953125, "learning_rate": 3.4175085134862128e-06, "loss": -0.0006, "num_tokens": 137424605.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.953125, "completions/max_length": 2048.0, "completions/max_terminated_length": 2006.0, "completions/mean_length": 654.0859375, "completions/mean_terminated_length": 646.8939208984375, "completions/min_length": 176.0, "completions/min_terminated_length": 176.0, "epoch": 0.19073203492276697, "frac_reward_zero_std": 0.96875, "grad_norm": 0.0655278626205979, "kl": 0.10150146484375, "learning_rate": 3.4035763488816953e-06, "loss": 0.0028, "num_tokens": 138080889.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1689.0, "completions/mean_length": 611.712890625, "completions/mean_terminated_length": 608.9021606445312, "completions/min_length": 195.0, "completions/min_terminated_length": 195.0, "epoch": 0.19162749048578465, "frac_reward_zero_std": 1.0, "grad_norm": 0.006893797084836839, "kl": 0.06103515625, "learning_rate": 3.3896167615225594e-06, "loss": 0.0006, "num_tokens": 138714822.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1959.0, "completions/mean_length": 565.845703125, "completions/mean_terminated_length": 562.9451904296875, "completions/min_length": 143.0, "completions/min_terminated_length": 143.0, "epoch": 0.19252294604880232, "frac_reward_zero_std": 0.96875, "grad_norm": 0.05962691311018318, "kl": 0.06121826171875, "learning_rate": 3.375630337125133e-06, "loss": 0.0004, "num_tokens": 139281447.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -7.0, "completions/max_length": 1919.0, "completions/max_terminated_length": 1919.0, "completions/mean_length": 620.06640625, "completions/mean_terminated_length": 620.06640625, "completions/min_length": 170.0, "completions/min_terminated_length": 170.0, "epoch": 0.19341840161182, "frac_reward_zero_std": 1.0, "grad_norm": 0.007079063914301492, "kl": 0.06048583984375, "learning_rate": 3.361617662531772e-06, "loss": 0.0006, "num_tokens": 139878761.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1923.0, "completions/mean_length": 627.935546875, "completions/mean_terminated_length": 622.36669921875, "completions/min_length": 98.0, "completions/min_terminated_length": 98.0, "epoch": 0.1943138571748377, "frac_reward_zero_std": 0.96875, "grad_norm": 0.05517179205307519, "kl": 0.05999755859375, "learning_rate": 3.347579325686237e-06, "loss": 0.0037, "num_tokens": 140503432.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1972.0, "completions/mean_length": 614.9609375, "completions/mean_terminated_length": 609.3411865234375, "completions/min_length": 138.0, "completions/min_terminated_length": 138.0, "epoch": 0.19520931273785538, "frac_reward_zero_std": 0.96875, "grad_norm": 0.06462999787985825, "kl": 0.06182861328125, "learning_rate": 3.333515915609027e-06, "loss": 0.0039, "num_tokens": 141108180.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -7.0, "completions/max_length": 1883.0, "completions/max_terminated_length": 1883.0, "completions/mean_length": 640.31640625, "completions/mean_terminated_length": 640.31640625, "completions/min_length": 144.0, "completions/min_terminated_length": 144.0, "epoch": 0.19610476830087306, "frac_reward_zero_std": 1.0, "grad_norm": 0.00663515657731032, "kl": 0.05987548828125, "learning_rate": 3.3194280223726616e-06, "loss": 0.0006, "num_tokens": 141779206.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 2013.0, "completions/mean_length": 654.919921875, "completions/mean_terminated_length": 649.4569091796875, "completions/min_length": 186.0, "completions/min_terminated_length": 186.0, "epoch": 0.19700022386389077, "frac_reward_zero_std": 0.90625, "grad_norm": 0.09335614343429051, "kl": 0.0611572265625, "learning_rate": 3.305316237076927e-06, "loss": -0.0058, "num_tokens": 142449213.0, "reward": 0.09921875596046448, "reward_std": 0.0026298905722796917, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.9921875, "rewards/format_reward/std": 0.08812850713729858, "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": -6.9375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1987.0, "completions/mean_length": 599.158203125, "completions/mean_terminated_length": 588.3779296875, "completions/min_length": 173.0, "completions/min_terminated_length": 173.0, "epoch": 0.19789567942690844, "frac_reward_zero_std": 0.96875, "grad_norm": 0.03817769249302888, "kl": 0.09942626953125, "learning_rate": 3.291181151824071e-06, "loss": 0.0014, "num_tokens": 143030670.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.953125, "completions/max_length": 2048.0, "completions/max_terminated_length": 1530.0, "completions/mean_length": 607.646484375, "completions/mean_terminated_length": 599.1571655273438, "completions/min_length": 198.0, "completions/min_terminated_length": 198.0, "epoch": 0.19879113498992612, "frac_reward_zero_std": 1.0, "grad_norm": 0.007676197980008601, "kl": 0.06060791015625, "learning_rate": 3.27702335969396e-06, "loss": 0.0006, "num_tokens": 143625449.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1841.0, "completions/mean_length": 659.720703125, "completions/mean_terminated_length": 637.6845703125, "completions/min_length": 186.0, "completions/min_terminated_length": 186.0, "epoch": 0.1996865905529438, "frac_reward_zero_std": 0.9375, "grad_norm": 0.0787303783620633, "kl": 0.09423828125, "learning_rate": 3.2628434547191985e-06, "loss": 0.0004, "num_tokens": 144259370.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "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": -6.953125, "completions/max_length": 2048.0, "completions/max_terminated_length": 1851.0, "completions/mean_length": 631.271484375, "completions/mean_terminated_length": 622.9214477539062, "completions/min_length": 186.0, "completions/min_terminated_length": 186.0, "epoch": 0.2005820461159615, "frac_reward_zero_std": 1.0, "grad_norm": 0.005402341436416679, "kl": 0.0595703125, "learning_rate": 3.2486420318601973e-06, "loss": 0.0006, "num_tokens": 144882757.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.890625, "completions/max_length": 2048.0, "completions/max_terminated_length": 1912.0, "completions/mean_length": 676.494140625, "completions/mean_terminated_length": 657.4832153320312, "completions/min_length": 157.0, "completions/min_terminated_length": 157.0, "epoch": 0.20147750167897918, "frac_reward_zero_std": 1.0, "grad_norm": 0.005625148130404837, "kl": 0.05645751953125, "learning_rate": 3.2344196869802187e-06, "loss": 0.0006, "num_tokens": 145555714.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.9375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1901.0, "completions/mean_length": 654.791015625, "completions/mean_terminated_length": 643.8208618164062, "completions/min_length": 214.0, "completions/min_terminated_length": 214.0, "epoch": 0.20237295724199686, "frac_reward_zero_std": 1.0, "grad_norm": 0.006171692676610506, "kl": 0.06085205078125, "learning_rate": 3.2201770168203694e-06, "loss": 0.0006, "num_tokens": 146156487.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.9375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1982.0, "completions/mean_length": 709.08203125, "completions/mean_terminated_length": 698.5393676757812, "completions/min_length": 185.0, "completions/min_terminated_length": 185.0, "epoch": 0.20326841280501456, "frac_reward_zero_std": 1.0, "grad_norm": 0.007539541625438744, "kl": 0.0589599609375, "learning_rate": 3.205914618974563e-06, "loss": 0.0006, "num_tokens": 146838753.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -7.0, "completions/max_length": 1851.0, "completions/max_terminated_length": 1851.0, "completions/mean_length": 648.322265625, "completions/mean_terminated_length": 648.322265625, "completions/min_length": 180.0, "completions/min_terminated_length": 180.0, "epoch": 0.20416386836803224, "frac_reward_zero_std": 0.90625, "grad_norm": 0.09389806559017558, "kl": 0.05657958984375, "learning_rate": 3.1916330918644496e-06, "loss": -0.0014, "num_tokens": 147471062.0, "reward": 0.09941406548023224, "reward_std": 0.0023437500931322575, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.994140625, "rewards/format_reward/std": 0.07639661431312561, "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1809.0, "completions/mean_length": 631.474609375, "completions/mean_terminated_length": 628.7025146484375, "completions/min_length": 163.0, "completions/min_terminated_length": 163.0, "epoch": 0.20505932393104992, "frac_reward_zero_std": 0.96875, "grad_norm": 0.045648453246142924, "kl": 0.0579833984375, "learning_rate": 3.177333034714303e-06, "loss": 0.0005, "num_tokens": 148087913.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.921875, "completions/max_length": 2048.0, "completions/max_terminated_length": 2032.0, "completions/mean_length": 649.294921875, "completions/mean_terminated_length": 635.5009765625, "completions/min_length": 217.0, "completions/min_terminated_length": 217.0, "epoch": 0.2059547794940676, "frac_reward_zero_std": 0.96875, "grad_norm": 0.05962054355895875, "kl": 0.058349609375, "learning_rate": 3.1630150475258813e-06, "loss": 0.0154, "num_tokens": 148741456.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -7.0, "completions/max_length": 1859.0, "completions/max_terminated_length": 1859.0, "completions/mean_length": 608.193359375, "completions/mean_terminated_length": 608.193359375, "completions/min_length": 168.0, "completions/min_terminated_length": 168.0, "epoch": 0.2068502350570853, "frac_reward_zero_std": 0.96875, "grad_norm": 0.049023036722908975, "kl": 0.05743408203125, "learning_rate": 3.148679731053252e-06, "loss": 0.0008, "num_tokens": 149318099.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.953125, "completions/max_length": 2048.0, "completions/max_terminated_length": 1938.0, "completions/mean_length": 625.474609375, "completions/mean_terminated_length": 617.0903930664062, "completions/min_length": 166.0, "completions/min_terminated_length": 166.0, "epoch": 0.20774569062010298, "frac_reward_zero_std": 0.9375, "grad_norm": 0.07269872169199251, "kl": 0.05938720703125, "learning_rate": 3.1343276867775805e-06, "loss": -0.0021, "num_tokens": 149926854.0, "reward": 0.09941406548023224, "reward_std": 0.0018486406188458204, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.994140625, "rewards/format_reward/std": 0.07639661431312561, "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1980.0, "completions/mean_length": 613.642578125, "completions/mean_terminated_length": 608.929443359375, "completions/min_length": 177.0, "completions/min_terminated_length": 177.0, "epoch": 0.20864114618312066, "frac_reward_zero_std": 0.90625, "grad_norm": 0.6311607973790101, "kl": 0.10137939453125, "learning_rate": 3.1199595168819043e-06, "loss": 0.0164, "num_tokens": 150555471.0, "reward": 0.09941406548023224, "reward_std": 0.0023437500931322575, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.994140625, "rewards/format_reward/std": 0.07639661431312561, "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": -6.890625, "completions/max_length": 2048.0, "completions/max_terminated_length": 1926.0, "completions/mean_length": 645.9453125, "completions/mean_terminated_length": 626.5109252929688, "completions/min_length": 156.0, "completions/min_terminated_length": 156.0, "epoch": 0.20953660174613833, "frac_reward_zero_std": 0.875, "grad_norm": 0.6851753963574233, "kl": 0.09783935546875, "learning_rate": 3.105575824225852e-06, "loss": 0.0114, "num_tokens": 151199779.0, "reward": 0.09921875596046448, "reward_std": 0.0031250000465661287, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.9921875, "rewards/format_reward/std": 0.08812850713729858, "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": -6.9375, "completions/max_length": 2048.0, "completions/max_terminated_length": 2006.0, "completions/mean_length": 680.703125, "completions/mean_terminated_length": 669.93701171875, "completions/min_length": 209.0, "completions/min_terminated_length": 209.0, "epoch": 0.21043205730915604, "frac_reward_zero_std": 1.0, "grad_norm": 0.008950782233121669, "kl": 0.05804443359375, "learning_rate": 3.091177212320363e-06, "loss": 0.0006, "num_tokens": 151869563.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1773.0, "completions/mean_length": 600.86328125, "completions/mean_terminated_length": 598.0313110351562, "completions/min_length": 131.0, "completions/min_terminated_length": 131.0, "epoch": 0.21132751287217372, "frac_reward_zero_std": 0.96875, "grad_norm": 0.06847164647361609, "kl": 0.05633544921875, "learning_rate": 3.0767642853023538e-06, "loss": 0.0001, "num_tokens": 152453125.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.953125, "completions/max_length": 2048.0, "completions/max_terminated_length": 1943.0, "completions/mean_length": 621.181640625, "completions/mean_terminated_length": 612.7720947265625, "completions/min_length": 136.0, "completions/min_terminated_length": 136.0, "epoch": 0.2122229684351914, "frac_reward_zero_std": 0.96875, "grad_norm": 0.055953444725904807, "kl": 0.05670166015625, "learning_rate": 3.062337647909376e-06, "loss": 0.0002, "num_tokens": 153060290.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1898.0, "completions/mean_length": 596.9453125, "completions/mean_terminated_length": 594.1056518554688, "completions/min_length": 147.0, "completions/min_terminated_length": 147.0, "epoch": 0.2131184239982091, "frac_reward_zero_std": 0.96875, "grad_norm": 0.05271028890825897, "kl": 0.0595703125, "learning_rate": 3.04789790545424e-06, "loss": 0.0018, "num_tokens": 153654278.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1867.0, "completions/mean_length": 626.3359375, "completions/mean_terminated_length": 620.7608032226562, "completions/min_length": 184.0, "completions/min_terminated_length": 184.0, "epoch": 0.21401387956122678, "frac_reward_zero_std": 1.0, "grad_norm": 0.00539813134920801, "kl": 0.056640625, "learning_rate": 3.033445663799621e-06, "loss": 0.0006, "num_tokens": 154274802.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1754.0, "completions/mean_length": 632.125, "completions/mean_terminated_length": 626.5725708007812, "completions/min_length": 184.0, "completions/min_terminated_length": 184.0, "epoch": 0.21490933512424445, "frac_reward_zero_std": 1.0, "grad_norm": 0.0055914288499768205, "kl": 0.05413818359375, "learning_rate": 3.018981529332633e-06, "loss": 0.0005, "num_tokens": 154885650.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "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": -6.9375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1966.0, "completions/mean_length": 654.580078125, "completions/mean_terminated_length": 644.25390625, "completions/min_length": 189.0, "completions/min_terminated_length": 189.0, "epoch": 0.21580479068726213, "frac_reward_zero_std": 0.96875, "grad_norm": 0.06858938507620503, "kl": 0.08441162109375, "learning_rate": 3.00450610893939e-06, "loss": 0.0039, "num_tokens": 155542603.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1936.0, "completions/mean_length": 604.103515625, "completions/mean_terminated_length": 598.4412231445312, "completions/min_length": 180.0, "completions/min_terminated_length": 180.0, "epoch": 0.21670024625027984, "frac_reward_zero_std": 0.96875, "grad_norm": 0.08157612254136043, "kl": 0.05743408203125, "learning_rate": 2.9900200099795396e-06, "loss": 0.0117, "num_tokens": 156160816.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 242 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 2035.0, "completions/mean_length": 573.150390625, "completions/mean_terminated_length": 570.26416015625, "completions/min_length": 109.0, "completions/min_terminated_length": 109.0, "epoch": 0.21759570181329752, "frac_reward_zero_std": 0.96875, "grad_norm": 0.06343258576945847, "kl": 0.05889892578125, "learning_rate": 2.9755238402607826e-06, "loss": -0.0, "num_tokens": 156762141.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 243 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1798.0, "completions/mean_length": 644.193359375, "completions/mean_terminated_length": 641.4461669921875, "completions/min_length": 177.0, "completions/min_terminated_length": 177.0, "epoch": 0.2184911573763152, "frac_reward_zero_std": 0.96875, "grad_norm": 0.06649190527686924, "kl": 0.056884765625, "learning_rate": 2.961018208013367e-06, "loss": 0.0051, "num_tokens": 157430304.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 244 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1929.0, "completions/mean_length": 661.330078125, "completions/mean_terminated_length": 655.8922119140625, "completions/min_length": 173.0, "completions/min_terminated_length": 173.0, "epoch": 0.2193866129393329, "frac_reward_zero_std": 1.0, "grad_norm": 0.0058189104851992055, "kl": 0.056396484375, "learning_rate": 2.9465037218645694e-06, "loss": 0.0006, "num_tokens": 158107769.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 245 }, { "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": -7.0, "completions/max_length": 1780.0, "completions/max_terminated_length": 1780.0, "completions/mean_length": 600.646484375, "completions/mean_terminated_length": 600.646484375, "completions/min_length": 102.0, "completions/min_terminated_length": 102.0, "epoch": 0.22028206850235058, "frac_reward_zero_std": 1.0, "grad_norm": 0.006009210073300697, "kl": 0.060791015625, "learning_rate": 2.9319809908131604e-06, "loss": 0.0006, "num_tokens": 158705812.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 246 }, { "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": -6.953125, "completions/max_length": 2048.0, "completions/max_terminated_length": 1928.0, "completions/mean_length": 621.87890625, "completions/mean_terminated_length": 613.4735107421875, "completions/min_length": 194.0, "completions/min_terminated_length": 194.0, "epoch": 0.22117752406536825, "frac_reward_zero_std": 1.0, "grad_norm": 0.005694321164073396, "kl": 0.05706787109375, "learning_rate": 2.917450624203847e-06, "loss": 0.0006, "num_tokens": 159333622.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 247 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1845.0, "completions/mean_length": 599.318359375, "completions/mean_terminated_length": 593.6372680664062, "completions/min_length": 137.0, "completions/min_terminated_length": 137.0, "epoch": 0.22207297962838593, "frac_reward_zero_std": 1.0, "grad_norm": 0.005349239849816427, "kl": 0.05615234375, "learning_rate": 2.9029132317017118e-06, "loss": 0.0006, "num_tokens": 159950761.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 248 }, { "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": -7.0, "completions/max_length": 1974.0, "completions/max_terminated_length": 1974.0, "completions/mean_length": 542.564453125, "completions/mean_terminated_length": 542.564453125, "completions/min_length": 171.0, "completions/min_terminated_length": 171.0, "epoch": 0.22296843519140364, "frac_reward_zero_std": 0.96875, "grad_norm": 0.06951417104945298, "kl": 0.05462646484375, "learning_rate": 2.888369423266629e-06, "loss": 0.001, "num_tokens": 160511178.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 249 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1705.0, "completions/mean_length": 573.044921875, "completions/mean_terminated_length": 570.1585083007812, "completions/min_length": 170.0, "completions/min_terminated_length": 170.0, "epoch": 0.2238638907544213, "frac_reward_zero_std": 1.0, "grad_norm": 0.006103601784565662, "kl": 0.05767822265625, "learning_rate": 2.8738198091276712e-06, "loss": 0.0006, "num_tokens": 161116001.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 250 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1891.0, "completions/mean_length": 566.23828125, "completions/mean_terminated_length": 563.3385620117188, "completions/min_length": 170.0, "completions/min_terminated_length": 170.0, "epoch": 0.224759346317439, "frac_reward_zero_std": 1.0, "grad_norm": 0.005820946741876905, "kl": 0.05841064453125, "learning_rate": 2.859264999757509e-06, "loss": 0.0006, "num_tokens": 161671451.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 251 }, { "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": -6.953125, "completions/max_length": 2048.0, "completions/max_terminated_length": 1879.0, "completions/mean_length": 609.8671875, "completions/mean_terminated_length": 604.1729125976562, "completions/min_length": 146.0, "completions/min_terminated_length": 146.0, "epoch": 0.2256548018804567, "frac_reward_zero_std": 0.9375, "grad_norm": 0.9391411364603389, "kl": 0.6685791015625, "learning_rate": 2.8447056058467928e-06, "loss": 0.0145, "num_tokens": 162310423.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "step": 252 }, { "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": -7.0, "completions/max_length": 1870.0, "completions/max_terminated_length": 1870.0, "completions/mean_length": 606.626953125, "completions/mean_terminated_length": 606.626953125, "completions/min_length": 162.0, "completions/min_terminated_length": 162.0, "epoch": 0.22655025744347437, "frac_reward_zero_std": 1.0, "grad_norm": 0.006495409905893996, "kl": 0.05908203125, "learning_rate": 2.830142238278531e-06, "loss": 0.0006, "num_tokens": 162959432.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 253 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 2048.0, "completions/mean_length": 617.728515625, "completions/mean_terminated_length": 612.11962890625, "completions/min_length": 175.0, "completions/min_terminated_length": 175.0, "epoch": 0.22744571300649205, "frac_reward_zero_std": 1.0, "grad_norm": 0.0073227247170333624, "kl": 0.05572509765625, "learning_rate": 2.81557550810246e-06, "loss": 0.0006, "num_tokens": 163563917.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 254 }, { "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": -7.0, "completions/max_length": 2001.0, "completions/max_terminated_length": 2001.0, "completions/mean_length": 586.6796875, "completions/mean_terminated_length": 586.6796875, "completions/min_length": 142.0, "completions/min_terminated_length": 142.0, "epoch": 0.22834116856950973, "frac_reward_zero_std": 1.0, "grad_norm": 0.00564245091070103, "kl": 0.05596923828125, "learning_rate": 2.8010060265094026e-06, "loss": 0.0006, "num_tokens": 164157049.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 255 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 2027.0, "completions/mean_length": 594.05078125, "completions/mean_terminated_length": 589.6470947265625, "completions/min_length": 157.0, "completions/min_terminated_length": 157.0, "epoch": 0.22923662413252743, "frac_reward_zero_std": 0.96875, "grad_norm": 0.05172954991761498, "kl": 0.056396484375, "learning_rate": 2.786434404805629e-06, "loss": 0.0022, "num_tokens": 164766163.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 256 }, { "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": -6.984375, "completions/max_length": 1935.0, "completions/max_terminated_length": 1935.0, "completions/mean_length": 606.1328125, "completions/mean_terminated_length": 604.4226684570312, "completions/min_length": 189.0, "completions/min_terminated_length": 189.0, "epoch": 0.2301320796955451, "frac_reward_zero_std": 1.0, "grad_norm": 0.03439221585128269, "kl": 0.06646728515625, "learning_rate": 2.771861254387199e-06, "loss": 0.0007, "num_tokens": 165420887.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 257 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1828.0, "completions/mean_length": 619.390625, "completions/mean_terminated_length": 613.7882690429688, "completions/min_length": 230.0, "completions/min_terminated_length": 230.0, "epoch": 0.2310275352585628, "frac_reward_zero_std": 1.0, "grad_norm": 0.005817974948231771, "kl": 0.05487060546875, "learning_rate": 2.7572871867143204e-06, "loss": 0.0005, "num_tokens": 166037263.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 258 }, { "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": -7.0, "completions/max_length": 2017.0, "completions/max_terminated_length": 2017.0, "completions/mean_length": 578.689453125, "completions/mean_terminated_length": 578.689453125, "completions/min_length": 207.0, "completions/min_terminated_length": 207.0, "epoch": 0.23192299082158047, "frac_reward_zero_std": 1.0, "grad_norm": 0.006405651142742982, "kl": 0.05810546875, "learning_rate": 2.742712813285681e-06, "loss": 0.0006, "num_tokens": 166660208.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 259 }, { "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": -6.953125, "completions/max_length": 2048.0, "completions/max_terminated_length": 1865.0, "completions/mean_length": 621.69921875, "completions/mean_terminated_length": 613.292724609375, "completions/min_length": 193.0, "completions/min_terminated_length": 193.0, "epoch": 0.23281844638459817, "frac_reward_zero_std": 0.96875, "grad_norm": 0.05751604346871723, "kl": 0.05535888671875, "learning_rate": 2.7281387456128017e-06, "loss": 0.0092, "num_tokens": 167327846.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 260 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1699.0, "completions/mean_length": 627.65234375, "completions/mean_terminated_length": 624.872802734375, "completions/min_length": 153.0, "completions/min_terminated_length": 153.0, "epoch": 0.23371390194761585, "frac_reward_zero_std": 1.0, "grad_norm": 0.005698567996800333, "kl": 0.05517578125, "learning_rate": 2.7135655951943716e-06, "loss": 0.0006, "num_tokens": 167989476.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 261 }, { "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": -7.0, "completions/max_length": 1988.0, "completions/max_terminated_length": 1988.0, "completions/mean_length": 594.044921875, "completions/mean_terminated_length": 594.044921875, "completions/min_length": 158.0, "completions/min_terminated_length": 158.0, "epoch": 0.23460935751063353, "frac_reward_zero_std": 1.0, "grad_norm": 0.005400704242655025, "kl": 0.0574951171875, "learning_rate": 2.698993973490598e-06, "loss": 0.0006, "num_tokens": 168635003.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 262 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1762.0, "completions/mean_length": 555.1796875, "completions/mean_terminated_length": 549.3255004882812, "completions/min_length": 174.0, "completions/min_terminated_length": 174.0, "epoch": 0.23550481307365123, "frac_reward_zero_std": 0.96875, "grad_norm": 0.08504139270976999, "kl": 0.05609130859375, "learning_rate": 2.6844244918975416e-06, "loss": 0.0133, "num_tokens": 169199639.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 263 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1924.0, "completions/mean_length": 585.38671875, "completions/mean_terminated_length": 582.5244750976562, "completions/min_length": 204.0, "completions/min_terminated_length": 204.0, "epoch": 0.2364002686366689, "frac_reward_zero_std": 0.96875, "grad_norm": 0.03248029309967152, "kl": 0.0560302734375, "learning_rate": 2.66985776172147e-06, "loss": 0.0153, "num_tokens": 169786493.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 264 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1704.0, "completions/mean_length": 587.125, "completions/mean_terminated_length": 584.26611328125, "completions/min_length": 190.0, "completions/min_terminated_length": 190.0, "epoch": 0.2372957241996866, "frac_reward_zero_std": 1.0, "grad_norm": 0.0054535619814076244, "kl": 0.0555419921875, "learning_rate": 2.6552943941532088e-06, "loss": 0.0006, "num_tokens": 170403005.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 265 }, { "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": -7.0, "completions/max_length": 2000.0, "completions/max_terminated_length": 2000.0, "completions/mean_length": 578.591796875, "completions/mean_terminated_length": 578.591796875, "completions/min_length": 131.0, "completions/min_terminated_length": 131.0, "epoch": 0.23819117976270426, "frac_reward_zero_std": 1.0, "grad_norm": 0.005800541107133192, "kl": 0.0567626953125, "learning_rate": 2.6407350002424927e-06, "loss": 0.0006, "num_tokens": 171022652.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 266 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1917.0, "completions/mean_length": 598.119140625, "completions/mean_terminated_length": 595.2817993164062, "completions/min_length": 103.0, "completions/min_terminated_length": 103.0, "epoch": 0.23908663532572197, "frac_reward_zero_std": 1.0, "grad_norm": 0.005198138170323681, "kl": 0.05340576171875, "learning_rate": 2.626180190872329e-06, "loss": 0.0005, "num_tokens": 171613625.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 267 }, { "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": -7.0, "completions/max_length": 1365.0, "completions/max_terminated_length": 1365.0, "completions/mean_length": 574.216796875, "completions/mean_terminated_length": 574.216796875, "completions/min_length": 152.0, "completions/min_terminated_length": 152.0, "epoch": 0.23998209088873965, "frac_reward_zero_std": 1.0, "grad_norm": 0.00526669304671622, "kl": 0.05548095703125, "learning_rate": 2.611630576733372e-06, "loss": 0.0006, "num_tokens": 172219752.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 268 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1925.0, "completions/mean_length": 590.83984375, "completions/mean_terminated_length": 587.98828125, "completions/min_length": 207.0, "completions/min_terminated_length": 207.0, "epoch": 0.24087754645175732, "frac_reward_zero_std": 1.0, "grad_norm": 0.005186941679267824, "kl": 0.0560302734375, "learning_rate": 2.5970867682982885e-06, "loss": 0.0006, "num_tokens": 172821350.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 269 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 2007.0, "completions/mean_length": 596.86328125, "completions/mean_terminated_length": 591.172607421875, "completions/min_length": 102.0, "completions/min_terminated_length": 102.0, "epoch": 0.24177300201477503, "frac_reward_zero_std": 1.0, "grad_norm": 0.005093847327967122, "kl": 0.05621337890625, "learning_rate": 2.582549375796154e-06, "loss": 0.0006, "num_tokens": 173464784.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 270 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1905.0, "completions/mean_length": 593.822265625, "completions/mean_terminated_length": 588.11962890625, "completions/min_length": 154.0, "completions/min_terminated_length": 154.0, "epoch": 0.2426684575777927, "frac_reward_zero_std": 0.96875, "grad_norm": 0.060575861215051076, "kl": 0.05645751953125, "learning_rate": 2.568019009186841e-06, "loss": -0.0023, "num_tokens": 174030725.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 271 }, { "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": -7.0, "completions/max_length": 1777.0, "completions/max_terminated_length": 1777.0, "completions/mean_length": 542.765625, "completions/mean_terminated_length": 542.765625, "completions/min_length": 214.0, "completions/min_terminated_length": 214.0, "epoch": 0.24356391314081038, "frac_reward_zero_std": 0.96875, "grad_norm": 0.04746305553350059, "kl": 0.0526123046875, "learning_rate": 2.5534962781354317e-06, "loss": -0.0031, "num_tokens": 174611821.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 272 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1549.0, "completions/mean_length": 583.73046875, "completions/mean_terminated_length": 577.98828125, "completions/min_length": 153.0, "completions/min_terminated_length": 153.0, "epoch": 0.24445936870382806, "frac_reward_zero_std": 1.0, "grad_norm": 0.0045173693793721685, "kl": 0.0533447265625, "learning_rate": 2.538981791986634e-06, "loss": 0.0005, "num_tokens": 175206915.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 273 }, { "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": -7.0, "completions/max_length": 2019.0, "completions/max_terminated_length": 2019.0, "completions/mean_length": 567.197265625, "completions/mean_terminated_length": 567.197265625, "completions/min_length": 106.0, "completions/min_terminated_length": 106.0, "epoch": 0.24535482426684577, "frac_reward_zero_std": 0.96875, "grad_norm": 0.06113247025112339, "kl": 0.05450439453125, "learning_rate": 2.524476159739218e-06, "loss": 0.0005, "num_tokens": 175822152.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 274 }, { "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": -7.0, "completions/max_length": 1987.0, "completions/max_terminated_length": 1987.0, "completions/mean_length": 585.8984375, "completions/mean_terminated_length": 585.8984375, "completions/min_length": 137.0, "completions/min_terminated_length": 137.0, "epoch": 0.24625027982986344, "frac_reward_zero_std": 1.0, "grad_norm": 0.00573175071522722, "kl": 0.0528564453125, "learning_rate": 2.5099799900204607e-06, "loss": 0.0005, "num_tokens": 176426356.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 275 }, { "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": -6.984375, "completions/max_length": 1939.0, "completions/max_terminated_length": 1939.0, "completions/mean_length": 591.904296875, "completions/mean_terminated_length": 589.8297119140625, "completions/min_length": 171.0, "completions/min_terminated_length": 171.0, "epoch": 0.24714573539288112, "frac_reward_zero_std": 1.0, "grad_norm": 0.006265698499194678, "kl": 0.0513916015625, "learning_rate": 2.4954938910606108e-06, "loss": 0.0005, "num_tokens": 177004131.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 276 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1694.0, "completions/mean_length": 543.6328125, "completions/mean_terminated_length": 540.6888427734375, "completions/min_length": 162.0, "completions/min_terminated_length": 162.0, "epoch": 0.24804119095589883, "frac_reward_zero_std": 0.96875, "grad_norm": 0.06559073582245414, "kl": 0.05224609375, "learning_rate": 2.481018470667368e-06, "loss": 0.0005, "num_tokens": 177565431.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 277 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1680.0, "completions/mean_length": 568.951171875, "completions/mean_terminated_length": 563.87255859375, "completions/min_length": 185.0, "completions/min_terminated_length": 185.0, "epoch": 0.2489366465189165, "frac_reward_zero_std": 0.90625, "grad_norm": 13.405018764621035, "kl": 0.37408447265625, "learning_rate": 2.4665543362003802e-06, "loss": 0.0184, "num_tokens": 178140382.0, "reward": 0.09941406548023224, "reward_std": 0.0023437500931322575, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.994140625, "rewards/format_reward/std": 0.07639661431312561, "step": 278 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1781.0, "completions/mean_length": 582.24609375, "completions/mean_terminated_length": 579.377685546875, "completions/min_length": 181.0, "completions/min_terminated_length": 181.0, "epoch": 0.24983210208193418, "frac_reward_zero_std": 0.96875, "grad_norm": 0.07793443113322279, "kl": 0.05450439453125, "learning_rate": 2.4521020945457615e-06, "loss": 0.0025, "num_tokens": 178751148.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 279 }, { "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": -7.0, "completions/max_length": 1761.0, "completions/max_terminated_length": 1761.0, "completions/mean_length": 555.86328125, "completions/mean_terminated_length": 555.86328125, "completions/min_length": 146.0, "completions/min_terminated_length": 146.0, "epoch": 0.25072755764495186, "frac_reward_zero_std": 1.0, "grad_norm": 0.005138666007838519, "kl": 0.05181884765625, "learning_rate": 2.4376623520906255e-06, "loss": 0.0005, "num_tokens": 179352646.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 280 }, { "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": -7.0, "completions/max_length": 1485.0, "completions/max_terminated_length": 1485.0, "completions/mean_length": 591.35546875, "completions/mean_terminated_length": 591.35546875, "completions/min_length": 207.0, "completions/min_terminated_length": 207.0, "epoch": 0.25162301320796954, "frac_reward_zero_std": 1.0, "grad_norm": 0.005950683891807377, "kl": 0.05291748046875, "learning_rate": 2.4232357146976478e-06, "loss": 0.0005, "num_tokens": 179970588.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 281 }, { "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": -6.984375, "completions/max_length": 1767.0, "completions/max_terminated_length": 1767.0, "completions/mean_length": 615.041015625, "completions/mean_terminated_length": 613.628173828125, "completions/min_length": 235.0, "completions/min_terminated_length": 235.0, "epoch": 0.2525184687709872, "frac_reward_zero_std": 1.0, "grad_norm": 24.033499681250188, "kl": 2.96002197265625, "learning_rate": 2.408822787679637e-06, "loss": 0.0295, "num_tokens": 180573601.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 282 }, { "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": -7.0, "completions/max_length": 1644.0, "completions/max_terminated_length": 1644.0, "completions/mean_length": 565.318359375, "completions/mean_terminated_length": 565.318359375, "completions/min_length": 178.0, "completions/min_terminated_length": 178.0, "epoch": 0.25341392433400495, "frac_reward_zero_std": 1.0, "grad_norm": 0.0051856058148775286, "kl": 0.0518798828125, "learning_rate": 2.3944241757741475e-06, "loss": 0.0005, "num_tokens": 181160180.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 283 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1993.0, "completions/mean_length": 617.09375, "completions/mean_terminated_length": 614.2935180664062, "completions/min_length": 182.0, "completions/min_terminated_length": 182.0, "epoch": 0.2543093798970226, "frac_reward_zero_std": 1.0, "grad_norm": 0.0068416300076160775, "kl": 0.05120849609375, "learning_rate": 2.380040483118097e-06, "loss": 0.0005, "num_tokens": 181755748.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 284 }, { "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": -7.0, "completions/max_length": 1894.0, "completions/max_terminated_length": 1894.0, "completions/mean_length": 583.033203125, "completions/mean_terminated_length": 583.033203125, "completions/min_length": 136.0, "completions/min_terminated_length": 136.0, "epoch": 0.2552048354600403, "frac_reward_zero_std": 0.9375, "grad_norm": 0.07301203946773449, "kl": 0.051513671875, "learning_rate": 2.365672313222419e-06, "loss": -0.0029, "num_tokens": 182358469.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "step": 285 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1728.0, "completions/mean_length": 628.9453125, "completions/mean_terminated_length": 623.3804321289062, "completions/min_length": 212.0, "completions/min_terminated_length": 212.0, "epoch": 0.256100291023058, "frac_reward_zero_std": 1.0, "grad_norm": 0.0054564530522672825, "kl": 0.05206298828125, "learning_rate": 2.351320268946749e-06, "loss": 0.0005, "num_tokens": 183015449.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 286 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1781.0, "completions/mean_length": 538.4453125, "completions/mean_terminated_length": 535.4912109375, "completions/min_length": 165.0, "completions/min_terminated_length": 165.0, "epoch": 0.25699574658607566, "frac_reward_zero_std": 1.0, "grad_norm": 0.005346092389195165, "kl": 0.0521240234375, "learning_rate": 2.336984952474119e-06, "loss": 0.0005, "num_tokens": 183558509.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 287 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1950.0, "completions/mean_length": 547.73046875, "completions/mean_terminated_length": 544.7944946289062, "completions/min_length": 145.0, "completions/min_terminated_length": 145.0, "epoch": 0.25789120214909333, "frac_reward_zero_std": 1.0, "grad_norm": 0.005394893401423969, "kl": 0.0499267578125, "learning_rate": 2.322666965285697e-06, "loss": 0.0005, "num_tokens": 184104819.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 288 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1585.0, "completions/mean_length": 588.498046875, "completions/mean_terminated_length": 582.7745361328125, "completions/min_length": 199.0, "completions/min_terminated_length": 199.0, "epoch": 0.258786657712111, "frac_reward_zero_std": 1.0, "grad_norm": 0.004969546618879694, "kl": 0.05023193359375, "learning_rate": 2.3083669081355507e-06, "loss": 0.0005, "num_tokens": 184713954.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 289 }, { "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": -7.0, "completions/max_length": 1976.0, "completions/max_terminated_length": 1976.0, "completions/mean_length": 552.771484375, "completions/mean_terminated_length": 552.771484375, "completions/min_length": 102.0, "completions/min_terminated_length": 102.0, "epoch": 0.25968211327512875, "frac_reward_zero_std": 1.0, "grad_norm": 0.008575669078842711, "kl": 0.0533447265625, "learning_rate": 2.2940853810254377e-06, "loss": 0.0005, "num_tokens": 185294381.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 290 }, { "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": -7.0, "completions/max_length": 1260.0, "completions/max_terminated_length": 1260.0, "completions/mean_length": 579.490234375, "completions/mean_terminated_length": 579.490234375, "completions/min_length": 226.0, "completions/min_terminated_length": 226.0, "epoch": 0.2605775688381464, "frac_reward_zero_std": 1.0, "grad_norm": 0.0052073739337190885, "kl": 0.05035400390625, "learning_rate": 2.2798229831796313e-06, "loss": 0.0005, "num_tokens": 185851064.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 291 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1751.0, "completions/mean_length": 630.0, "completions/mean_terminated_length": 627.2250366210938, "completions/min_length": 162.0, "completions/min_terminated_length": 162.0, "epoch": 0.2614730244011641, "frac_reward_zero_std": 0.96875, "grad_norm": 0.053543315977731425, "kl": 0.0494384765625, "learning_rate": 2.2655803130197816e-06, "loss": -0.0003, "num_tokens": 186467848.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 292 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1816.0, "completions/mean_length": 568.5859375, "completions/mean_terminated_length": 565.6907958984375, "completions/min_length": 189.0, "completions/min_terminated_length": 189.0, "epoch": 0.2623684799641818, "frac_reward_zero_std": 1.0, "grad_norm": 0.004634463449738635, "kl": 0.05108642578125, "learning_rate": 2.2513579681398034e-06, "loss": 0.0005, "num_tokens": 187048996.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 293 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1966.0, "completions/mean_length": 554.080078125, "completions/mean_terminated_length": 548.2216186523438, "completions/min_length": 135.0, "completions/min_terminated_length": 135.0, "epoch": 0.26326393552719946, "frac_reward_zero_std": 1.0, "grad_norm": 0.004680081807285644, "kl": 0.0513916015625, "learning_rate": 2.237156545280803e-06, "loss": 0.0005, "num_tokens": 187595901.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 294 }, { "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": -7.0, "completions/max_length": 1857.0, "completions/max_terminated_length": 1857.0, "completions/mean_length": 592.625, "completions/mean_terminated_length": 592.625, "completions/min_length": 158.0, "completions/min_terminated_length": 158.0, "epoch": 0.26415939109021713, "frac_reward_zero_std": 1.0, "grad_norm": 0.004679039787165404, "kl": 0.052001953125, "learning_rate": 2.2229766403060403e-06, "loss": 0.0005, "num_tokens": 188187341.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 295 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1792.0, "completions/mean_length": 614.19140625, "completions/mean_terminated_length": 610.5235595703125, "completions/min_length": 143.0, "completions/min_terminated_length": 143.0, "epoch": 0.2650548466532348, "frac_reward_zero_std": 0.96875, "grad_norm": 0.052447372839830086, "kl": 0.05328369140625, "learning_rate": 2.2088188481759305e-06, "loss": 0.0126, "num_tokens": 188791503.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 296 }, { "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": -6.984375, "completions/max_length": 1818.0, "completions/max_terminated_length": 1818.0, "completions/mean_length": 627.5078125, "completions/mean_terminated_length": 626.383544921875, "completions/min_length": 225.0, "completions/min_terminated_length": 225.0, "epoch": 0.26595030221625254, "frac_reward_zero_std": 0.96875, "grad_norm": 0.04957490670885095, "kl": 0.05078125, "learning_rate": 2.194683762923073e-06, "loss": -0.0005, "num_tokens": 189428755.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 297 }, { "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": -7.0, "completions/max_length": 1568.0, "completions/max_terminated_length": 1568.0, "completions/mean_length": 558.3046875, "completions/mean_terminated_length": 558.3046875, "completions/min_length": 157.0, "completions/min_terminated_length": 157.0, "epoch": 0.2668457577792702, "frac_reward_zero_std": 0.9375, "grad_norm": 0.07521000733044311, "kl": 0.05029296875, "learning_rate": 2.1805719776273387e-06, "loss": 0.0034, "num_tokens": 189996799.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "step": 298 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1702.0, "completions/mean_length": 595.80078125, "completions/mean_terminated_length": 590.1058959960938, "completions/min_length": 159.0, "completions/min_terminated_length": 159.0, "epoch": 0.2677412133422879, "frac_reward_zero_std": 0.9375, "grad_norm": 52.29210967783974, "kl": 2.88189697265625, "learning_rate": 2.166484084390974e-06, "loss": 0.0381, "num_tokens": 190597097.0, "reward": 0.09941406548023224, "reward_std": 0.0018486406188458204, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.994140625, "rewards/format_reward/std": 0.07639661431312561, "step": 299 }, { "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": -7.0, "completions/max_length": 1919.0, "completions/max_terminated_length": 1919.0, "completions/mean_length": 575.779296875, "completions/mean_terminated_length": 575.779296875, "completions/min_length": 179.0, "completions/min_terminated_length": 179.0, "epoch": 0.2686366689053056, "frac_reward_zero_std": 0.96875, "grad_norm": 0.05666066912224791, "kl": 0.0506591796875, "learning_rate": 2.1524206743137636e-06, "loss": 0.0024, "num_tokens": 191182056.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 300 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1711.0, "completions/mean_length": 602.306640625, "completions/mean_terminated_length": 599.4774780273438, "completions/min_length": 183.0, "completions/min_terminated_length": 183.0, "epoch": 0.26953212446832325, "frac_reward_zero_std": 0.96875, "grad_norm": 0.04902480949298609, "kl": 0.0518798828125, "learning_rate": 2.1383823374682287e-06, "loss": -0.0009, "num_tokens": 191805093.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 301 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1448.0, "completions/mean_length": 600.71484375, "completions/mean_terminated_length": 595.0392456054688, "completions/min_length": 198.0, "completions/min_terminated_length": 198.0, "epoch": 0.27042758003134093, "frac_reward_zero_std": 1.0, "grad_norm": 0.004860988243453146, "kl": 0.05169677734375, "learning_rate": 2.124369662874868e-06, "loss": 0.0005, "num_tokens": 192417747.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 302 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1668.0, "completions/mean_length": 565.7265625, "completions/mean_terminated_length": 562.8258056640625, "completions/min_length": 111.0, "completions/min_terminated_length": 111.0, "epoch": 0.2713230355943586, "frac_reward_zero_std": 0.96875, "grad_norm": 0.054741263922224406, "kl": 0.05029296875, "learning_rate": 2.110383238477441e-06, "loss": -0.0018, "num_tokens": 192973031.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 303 }, { "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": -6.984375, "completions/max_length": 1812.0, "completions/max_terminated_length": 1812.0, "completions/mean_length": 627.341796875, "completions/mean_terminated_length": 625.0234985351562, "completions/min_length": 166.0, "completions/min_terminated_length": 166.0, "epoch": 0.27221849115737634, "frac_reward_zero_std": 0.96875, "grad_norm": 1.1739951576785839, "kl": 0.18341064453125, "learning_rate": 2.096423651118305e-06, "loss": 0.0074, "num_tokens": 193610566.0, "reward": 0.099609375, "reward_std": 0.0010673906654119492, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "step": 304 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1740.0, "completions/mean_length": 594.041015625, "completions/mean_terminated_length": 591.1956787109375, "completions/min_length": 205.0, "completions/min_terminated_length": 205.0, "epoch": 0.273113946720394, "frac_reward_zero_std": 1.0, "grad_norm": 0.005936683963889181, "kl": 0.0511474609375, "learning_rate": 2.082491486513788e-06, "loss": 0.0005, "num_tokens": 194198091.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 305 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1528.0, "completions/mean_length": 582.513671875, "completions/mean_terminated_length": 579.6458129882812, "completions/min_length": 171.0, "completions/min_terminated_length": 171.0, "epoch": 0.2740094022834117, "frac_reward_zero_std": 1.0, "grad_norm": 0.006122848959639277, "kl": 0.05328369140625, "learning_rate": 2.0685873292296116e-06, "loss": 0.0005, "num_tokens": 194772946.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 306 }, { "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": -7.0, "completions/max_length": 1602.0, "completions/max_terminated_length": 1602.0, "completions/mean_length": 591.427734375, "completions/mean_terminated_length": 591.427734375, "completions/min_length": 166.0, "completions/min_terminated_length": 166.0, "epoch": 0.2749048578464294, "frac_reward_zero_std": 0.96875, "grad_norm": 0.05947760191798018, "kl": 0.05084228515625, "learning_rate": 2.054711762656369e-06, "loss": 0.0041, "num_tokens": 195386189.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 307 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1716.0, "completions/mean_length": 598.970703125, "completions/mean_terminated_length": 596.135009765625, "completions/min_length": 217.0, "completions/min_terminated_length": 217.0, "epoch": 0.27580031340944705, "frac_reward_zero_std": 0.9375, "grad_norm": 0.10573090050455233, "kl": 0.0560302734375, "learning_rate": 2.040865368985044e-06, "loss": 0.0109, "num_tokens": 195976750.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "step": 308 }, { "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": -6.9375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1736.0, "completions/mean_length": 646.505859375, "completions/mean_terminated_length": 635.470458984375, "completions/min_length": 162.0, "completions/min_terminated_length": 162.0, "epoch": 0.27669576897246473, "frac_reward_zero_std": 1.0, "grad_norm": 0.005228529652921396, "kl": 0.05120849609375, "learning_rate": 2.027048729182583e-06, "loss": 0.0005, "num_tokens": 196618753.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 309 }, { "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": -7.0, "completions/max_length": 1819.0, "completions/max_terminated_length": 1819.0, "completions/mean_length": 606.69921875, "completions/mean_terminated_length": 606.69921875, "completions/min_length": 149.0, "completions/min_terminated_length": 149.0, "epoch": 0.2775912245354824, "frac_reward_zero_std": 1.0, "grad_norm": 0.0051559766107908914, "kl": 0.05517578125, "learning_rate": 2.0132624229675205e-06, "loss": 0.0006, "num_tokens": 197266071.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 310 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1437.0, "completions/mean_length": 596.806640625, "completions/mean_terminated_length": 593.9667358398438, "completions/min_length": 201.0, "completions/min_terminated_length": 201.0, "epoch": 0.27848668009850014, "frac_reward_zero_std": 0.96875, "grad_norm": 0.0479412945784682, "kl": 0.05206298828125, "learning_rate": 1.9995070287856546e-06, "loss": 0.0012, "num_tokens": 197858996.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 311 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1617.0, "completions/mean_length": 578.19140625, "completions/mean_terminated_length": 575.3150634765625, "completions/min_length": 186.0, "completions/min_terminated_length": 186.0, "epoch": 0.2793821356615178, "frac_reward_zero_std": 0.96875, "grad_norm": 0.05124512126982719, "kl": 0.05169677734375, "learning_rate": 1.985783123785774e-06, "loss": 0.0183, "num_tokens": 198438822.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 312 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1749.0, "completions/mean_length": 568.921875, "completions/mean_terminated_length": 566.0274047851562, "completions/min_length": 149.0, "completions/min_terminated_length": 149.0, "epoch": 0.2802775912245355, "frac_reward_zero_std": 0.96875, "grad_norm": 0.05819859258148276, "kl": 0.0533447265625, "learning_rate": 1.9720912837954486e-06, "loss": -0.0, "num_tokens": 199030926.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 313 }, { "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": -7.0, "completions/max_length": 1491.0, "completions/max_terminated_length": 1491.0, "completions/mean_length": 554.88671875, "completions/mean_terminated_length": 554.88671875, "completions/min_length": 157.0, "completions/min_terminated_length": 157.0, "epoch": 0.28117304678755317, "frac_reward_zero_std": 0.9375, "grad_norm": 0.09616545890196271, "kl": 0.0570068359375, "learning_rate": 1.958432083296862e-06, "loss": 0.0083, "num_tokens": 199609012.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "step": 314 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1908.0, "completions/mean_length": 562.427734375, "completions/mean_terminated_length": 559.5205688476562, "completions/min_length": 138.0, "completions/min_terminated_length": 138.0, "epoch": 0.28206850235057085, "frac_reward_zero_std": 1.0, "grad_norm": 0.006128274293792745, "kl": 0.052490234375, "learning_rate": 1.9448060954027093e-06, "loss": 0.0005, "num_tokens": 200170303.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 315 }, { "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": -7.0, "completions/max_length": 1877.0, "completions/max_terminated_length": 1877.0, "completions/mean_length": 551.5390625, "completions/mean_terminated_length": 551.5390625, "completions/min_length": 103.0, "completions/min_terminated_length": 103.0, "epoch": 0.2829639579135885, "frac_reward_zero_std": 0.96875, "grad_norm": 0.06205881843423876, "kl": 0.0543212890625, "learning_rate": 1.931213891832153e-06, "loss": -0.0005, "num_tokens": 200752611.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 316 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1282.0, "completions/mean_length": 569.83984375, "completions/mean_terminated_length": 564.0431518554688, "completions/min_length": 131.0, "completions/min_terminated_length": 131.0, "epoch": 0.2838594134766062, "frac_reward_zero_std": 0.9375, "grad_norm": 0.07968967539050426, "kl": 0.0521240234375, "learning_rate": 1.9176560428868336e-06, "loss": 0.0085, "num_tokens": 201334673.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "step": 317 }, { "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": -7.0, "completions/max_length": 1782.0, "completions/max_terminated_length": 1782.0, "completions/mean_length": 618.275390625, "completions/mean_terminated_length": 618.275390625, "completions/min_length": 222.0, "completions/min_terminated_length": 222.0, "epoch": 0.2847548690396239, "frac_reward_zero_std": 1.0, "grad_norm": 0.005425161560628047, "kl": 0.054443359375, "learning_rate": 1.9041331174269373e-06, "loss": 0.0005, "num_tokens": 201992350.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 318 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1876.0, "completions/mean_length": 564.7734375, "completions/mean_terminated_length": 561.870849609375, "completions/min_length": 144.0, "completions/min_terminated_length": 144.0, "epoch": 0.2856503246026416, "frac_reward_zero_std": 1.0, "grad_norm": 0.005142824320868143, "kl": 0.0537109375, "learning_rate": 1.8906456828473341e-06, "loss": 0.0005, "num_tokens": 202598010.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 319 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1676.0, "completions/mean_length": 582.58984375, "completions/mean_terminated_length": 576.8432006835938, "completions/min_length": 140.0, "completions/min_terminated_length": 140.0, "epoch": 0.2865457801656593, "frac_reward_zero_std": 1.0, "grad_norm": 0.009530691407977925, "kl": 0.0511474609375, "learning_rate": 1.8771943050537656e-06, "loss": 0.0005, "num_tokens": 203186424.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 320 }, { "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": -7.0, "completions/max_length": 1453.0, "completions/max_terminated_length": 1453.0, "completions/mean_length": 580.2578125, "completions/mean_terminated_length": 580.2578125, "completions/min_length": 169.0, "completions/min_terminated_length": 169.0, "epoch": 0.28744123572867697, "frac_reward_zero_std": 1.0, "grad_norm": 0.005597161597226462, "kl": 0.05303955078125, "learning_rate": 1.8637795484391046e-06, "loss": 0.0005, "num_tokens": 203807692.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 321 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1518.0, "completions/mean_length": 560.55859375, "completions/mean_terminated_length": 556.5628051757812, "completions/min_length": 190.0, "completions/min_terminated_length": 190.0, "epoch": 0.28833669129169465, "frac_reward_zero_std": 1.0, "grad_norm": 10.190425237517521, "kl": 2.46160888671875, "learning_rate": 1.8504019758596698e-06, "loss": 0.0244, "num_tokens": 204418266.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 322 }, { "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": -7.0, "completions/max_length": 1940.0, "completions/max_terminated_length": 1940.0, "completions/mean_length": 554.6328125, "completions/mean_terminated_length": 554.6328125, "completions/min_length": 104.0, "completions/min_terminated_length": 104.0, "epoch": 0.2892321468547123, "frac_reward_zero_std": 1.0, "grad_norm": 0.007024406654251954, "kl": 0.0526123046875, "learning_rate": 1.8370621486116163e-06, "loss": 0.0005, "num_tokens": 205002910.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 323 }, { "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": -7.0, "completions/max_length": 1875.0, "completions/max_terminated_length": 1875.0, "completions/mean_length": 582.15625, "completions/mean_terminated_length": 582.15625, "completions/min_length": 156.0, "completions/min_terminated_length": 156.0, "epoch": 0.29012760241773, "frac_reward_zero_std": 1.0, "grad_norm": 0.007452427305034759, "kl": 0.0511474609375, "learning_rate": 1.823760626407377e-06, "loss": 0.0005, "num_tokens": 205639790.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 324 }, { "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": -7.0, "completions/max_length": 1902.0, "completions/max_terminated_length": 1902.0, "completions/mean_length": 606.71484375, "completions/mean_terminated_length": 606.71484375, "completions/min_length": 177.0, "completions/min_terminated_length": 177.0, "epoch": 0.2910230579807477, "frac_reward_zero_std": 1.0, "grad_norm": 0.005388949503012748, "kl": 0.0504150390625, "learning_rate": 1.8104979673521838e-06, "loss": 0.0005, "num_tokens": 206277644.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 325 }, { "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": -7.0, "completions/max_length": 1746.0, "completions/max_terminated_length": 1746.0, "completions/mean_length": 566.4375, "completions/mean_terminated_length": 566.4375, "completions/min_length": 154.0, "completions/min_terminated_length": 154.0, "epoch": 0.2919185135437654, "frac_reward_zero_std": 0.96875, "grad_norm": 0.06290666297265979, "kl": 0.05303955078125, "learning_rate": 1.7972747279206482e-06, "loss": -0.0001, "num_tokens": 206868460.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 326 }, { "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": -7.0, "completions/max_length": 1924.0, "completions/max_terminated_length": 1924.0, "completions/mean_length": 587.607421875, "completions/mean_terminated_length": 587.607421875, "completions/min_length": 177.0, "completions/min_terminated_length": 177.0, "epoch": 0.2928139691067831, "frac_reward_zero_std": 1.0, "grad_norm": 0.005627099408001751, "kl": 0.05084228515625, "learning_rate": 1.7840914629334122e-06, "loss": 0.0005, "num_tokens": 207492451.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 327 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1831.0, "completions/mean_length": 575.02734375, "completions/mean_terminated_length": 572.1448364257812, "completions/min_length": 169.0, "completions/min_terminated_length": 169.0, "epoch": 0.29370942466980077, "frac_reward_zero_std": 1.0, "grad_norm": 0.006024866616452142, "kl": 0.05047607421875, "learning_rate": 1.7709487255338731e-06, "loss": 0.0005, "num_tokens": 208115505.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 328 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1685.0, "completions/mean_length": 610.88671875, "completions/mean_terminated_length": 605.2510375976562, "completions/min_length": 130.0, "completions/min_terminated_length": 130.0, "epoch": 0.29460488023281844, "frac_reward_zero_std": 1.0, "grad_norm": 0.00783507654646286, "kl": 0.0528564453125, "learning_rate": 1.7578470671649684e-06, "loss": 0.0005, "num_tokens": 208772103.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 329 }, { "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": -7.0, "completions/max_length": 1555.0, "completions/max_terminated_length": 1555.0, "completions/mean_length": 544.369140625, "completions/mean_terminated_length": 544.369140625, "completions/min_length": 148.0, "completions/min_terminated_length": 148.0, "epoch": 0.2955003357958361, "frac_reward_zero_std": 0.96875, "grad_norm": 0.08261649735534114, "kl": 0.0511474609375, "learning_rate": 1.744787037546045e-06, "loss": 0.0142, "num_tokens": 209375716.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 330 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1732.0, "completions/mean_length": 595.642578125, "completions/mean_terminated_length": 589.9470825195312, "completions/min_length": 180.0, "completions/min_terminated_length": 180.0, "epoch": 0.2963957913588538, "frac_reward_zero_std": 0.96875, "grad_norm": 0.05439911125987834, "kl": 0.049560546875, "learning_rate": 1.731769184649788e-06, "loss": 0.0093, "num_tokens": 210026557.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 331 }, { "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": -7.0, "completions/max_length": 1854.0, "completions/max_terminated_length": 1854.0, "completions/mean_length": 576.931640625, "completions/mean_terminated_length": 576.931640625, "completions/min_length": 108.0, "completions/min_terminated_length": 108.0, "epoch": 0.2972912469218715, "frac_reward_zero_std": 0.96875, "grad_norm": 0.03824221393460093, "kl": 0.0489501953125, "learning_rate": 1.7187940546792325e-06, "loss": -0.0039, "num_tokens": 210615114.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 332 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1659.0, "completions/mean_length": 583.22265625, "completions/mean_terminated_length": 577.4784545898438, "completions/min_length": 139.0, "completions/min_terminated_length": 139.0, "epoch": 0.2981867024848892, "frac_reward_zero_std": 1.0, "grad_norm": 0.00502000048821662, "kl": 0.051513671875, "learning_rate": 1.7058621920448465e-06, "loss": 0.0005, "num_tokens": 211220828.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 333 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1862.0, "completions/mean_length": 604.205078125, "completions/mean_terminated_length": 601.379638671875, "completions/min_length": 191.0, "completions/min_terminated_length": 191.0, "epoch": 0.2990821580479069, "frac_reward_zero_std": 1.0, "grad_norm": 0.009980348313441741, "kl": 0.0511474609375, "learning_rate": 1.6929741393416855e-06, "loss": 0.0005, "num_tokens": 211842757.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 334 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1599.0, "completions/mean_length": 558.705078125, "completions/mean_terminated_length": 555.7905883789062, "completions/min_length": 132.0, "completions/min_terminated_length": 132.0, "epoch": 0.29997761361092457, "frac_reward_zero_std": 1.0, "grad_norm": 0.005398017340368072, "kl": 0.05047607421875, "learning_rate": 1.6801304373266286e-06, "loss": 0.0005, "num_tokens": 212409118.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 335 }, { "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": -6.984375, "completions/max_length": 1582.0, "completions/max_terminated_length": 1582.0, "completions/mean_length": 577.64453125, "completions/mean_terminated_length": 575.9510498046875, "completions/min_length": 170.0, "completions/min_terminated_length": 170.0, "epoch": 0.30087306917394224, "frac_reward_zero_std": 0.96875, "grad_norm": 0.0400953710083368, "kl": 0.0546875, "learning_rate": 1.667331624895689e-06, "loss": -0.0065, "num_tokens": 213017944.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 336 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1681.0, "completions/mean_length": 585.810546875, "completions/mean_terminated_length": 582.9490966796875, "completions/min_length": 152.0, "completions/min_terminated_length": 152.0, "epoch": 0.3017685247369599, "frac_reward_zero_std": 1.0, "grad_norm": 0.005855989237846307, "kl": 0.05377197265625, "learning_rate": 1.6545782390614037e-06, "loss": 0.0005, "num_tokens": 213613735.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 337 }, { "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": -6.984375, "completions/max_length": 1459.0, "completions/max_terminated_length": 1459.0, "completions/mean_length": 552.6015625, "completions/mean_terminated_length": 551.0430297851562, "completions/min_length": 129.0, "completions/min_terminated_length": 129.0, "epoch": 0.3026639802999776, "frac_reward_zero_std": 1.0, "grad_norm": 0.021342755030715328, "kl": 0.0576171875, "learning_rate": 1.6418708149302992e-06, "loss": 0.0006, "num_tokens": 214163835.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 338 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1754.0, "completions/mean_length": 568.935546875, "completions/mean_terminated_length": 566.0410766601562, "completions/min_length": 145.0, "completions/min_terminated_length": 145.0, "epoch": 0.3035594358629953, "frac_reward_zero_std": 1.0, "grad_norm": 0.004817435169613062, "kl": 0.05023193359375, "learning_rate": 1.6292098856804423e-06, "loss": 0.0005, "num_tokens": 214752170.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 339 }, { "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": -6.953125, "completions/max_length": 2048.0, "completions/max_terminated_length": 1948.0, "completions/mean_length": 632.53125, "completions/mean_terminated_length": 624.1885986328125, "completions/min_length": 163.0, "completions/min_terminated_length": 163.0, "epoch": 0.304454891426013, "frac_reward_zero_std": 0.96875, "grad_norm": 0.042312900898714305, "kl": 0.05169677734375, "learning_rate": 1.6165959825390661e-06, "loss": -0.0005, "num_tokens": 215456714.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 340 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1992.0, "completions/mean_length": 611.076171875, "completions/mean_terminated_length": 605.4412231445312, "completions/min_length": 200.0, "completions/min_terminated_length": 200.0, "epoch": 0.3053503469890307, "frac_reward_zero_std": 0.96875, "grad_norm": 0.06514871663082826, "kl": 0.0489501953125, "learning_rate": 1.604029634760284e-06, "loss": 0.0068, "num_tokens": 216074449.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 341 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1917.0, "completions/mean_length": 607.349609375, "completions/mean_terminated_length": 604.5303344726562, "completions/min_length": 147.0, "completions/min_terminated_length": 147.0, "epoch": 0.30624580255204836, "frac_reward_zero_std": 1.0, "grad_norm": 0.004697869051960354, "kl": 0.04931640625, "learning_rate": 1.59151136960288e-06, "loss": 0.0005, "num_tokens": 216714900.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 342 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1952.0, "completions/mean_length": 585.927734375, "completions/mean_terminated_length": 583.0665283203125, "completions/min_length": 154.0, "completions/min_terminated_length": 154.0, "epoch": 0.30714125811506604, "frac_reward_zero_std": 0.96875, "grad_norm": 0.041363506025014346, "kl": 0.049560546875, "learning_rate": 1.5790417123081903e-06, "loss": -0.0014, "num_tokens": 217293791.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 343 }, { "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": -6.984375, "completions/max_length": 1563.0, "completions/max_terminated_length": 1563.0, "completions/mean_length": 552.130859375, "completions/mean_terminated_length": 550.9921875, "completions/min_length": 151.0, "completions/min_terminated_length": 151.0, "epoch": 0.3080367136780837, "frac_reward_zero_std": 1.0, "grad_norm": 0.024577605367231227, "kl": 0.05859375, "learning_rate": 1.5666211860780583e-06, "loss": 0.0006, "num_tokens": 217848418.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 344 }, { "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": -7.0, "completions/max_length": 1901.0, "completions/max_terminated_length": 1901.0, "completions/mean_length": 602.701171875, "completions/mean_terminated_length": 602.701171875, "completions/min_length": 197.0, "completions/min_terminated_length": 197.0, "epoch": 0.3089321692411014, "frac_reward_zero_std": 1.0, "grad_norm": 0.004385742088872913, "kl": 0.04888916015625, "learning_rate": 1.5542503120528918e-06, "loss": 0.0005, "num_tokens": 218470825.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 345 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1693.0, "completions/mean_length": 612.8984375, "completions/mean_terminated_length": 607.2706298828125, "completions/min_length": 200.0, "completions/min_terminated_length": 200.0, "epoch": 0.3098276248041191, "frac_reward_zero_std": 1.0, "grad_norm": 0.004942302367708638, "kl": 0.04833984375, "learning_rate": 1.5419296092897866e-06, "loss": 0.0005, "num_tokens": 219057701.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 346 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1459.0, "completions/mean_length": 554.208984375, "completions/mean_terminated_length": 551.2857055664062, "completions/min_length": 156.0, "completions/min_terminated_length": 156.0, "epoch": 0.3107230803671368, "frac_reward_zero_std": 1.0, "grad_norm": 0.004833275143594728, "kl": 0.0494384765625, "learning_rate": 1.529659594740755e-06, "loss": 0.0005, "num_tokens": 219618176.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 347 }, { "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": -7.0, "completions/max_length": 1726.0, "completions/max_terminated_length": 1726.0, "completions/mean_length": 615.357421875, "completions/mean_terminated_length": 615.357421875, "completions/min_length": 171.0, "completions/min_terminated_length": 171.0, "epoch": 0.3116185359301545, "frac_reward_zero_std": 0.96875, "grad_norm": 0.0541376938239617, "kl": 0.0521240234375, "learning_rate": 1.5174407832310338e-06, "loss": -0.0004, "num_tokens": 220263719.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 348 }, { "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": -7.0, "completions/max_length": 1533.0, "completions/max_terminated_length": 1533.0, "completions/mean_length": 606.06640625, "completions/mean_terminated_length": 606.06640625, "completions/min_length": 215.0, "completions/min_terminated_length": 215.0, "epoch": 0.31251399149317216, "frac_reward_zero_std": 1.0, "grad_norm": 0.007838170301132085, "kl": 0.04876708984375, "learning_rate": 1.5052736874374815e-06, "loss": 0.0005, "num_tokens": 220890633.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 349 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1796.0, "completions/mean_length": 563.53125, "completions/mean_terminated_length": 557.7098388671875, "completions/min_length": 177.0, "completions/min_terminated_length": 177.0, "epoch": 0.31340944705618984, "frac_reward_zero_std": 1.0, "grad_norm": 0.0041665151315056406, "kl": 0.04742431640625, "learning_rate": 1.4931588178670695e-06, "loss": 0.0005, "num_tokens": 221464633.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 350 }, { "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": -7.0, "completions/max_length": 1522.0, "completions/max_terminated_length": 1522.0, "completions/mean_length": 596.26953125, "completions/mean_terminated_length": 596.26953125, "completions/min_length": 218.0, "completions/min_terminated_length": 218.0, "epoch": 0.3143049026192075, "frac_reward_zero_std": 1.0, "grad_norm": 0.004236470661639679, "kl": 0.0465087890625, "learning_rate": 1.4810966828354605e-06, "loss": 0.0005, "num_tokens": 222123811.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 351 }, { "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": -7.0, "completions/max_length": 1732.0, "completions/max_terminated_length": 1732.0, "completions/mean_length": 579.677734375, "completions/mean_terminated_length": 579.677734375, "completions/min_length": 107.0, "completions/min_terminated_length": 107.0, "epoch": 0.3152003581822252, "frac_reward_zero_std": 1.0, "grad_norm": 0.004236180586681027, "kl": 0.0506591796875, "learning_rate": 1.469087788445684e-06, "loss": 0.0005, "num_tokens": 222724574.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 352 }, { "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": -7.0, "completions/max_length": 1603.0, "completions/max_terminated_length": 1603.0, "completions/mean_length": 563.109375, "completions/mean_terminated_length": 563.109375, "completions/min_length": 191.0, "completions/min_terminated_length": 191.0, "epoch": 0.31609581374524287, "frac_reward_zero_std": 1.0, "grad_norm": 0.004163722780438454, "kl": 0.04901123046875, "learning_rate": 1.4571326385668965e-06, "loss": 0.0005, "num_tokens": 223341510.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 353 }, { "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": -7.0, "completions/max_length": 1976.0, "completions/max_terminated_length": 1976.0, "completions/mean_length": 594.8515625, "completions/mean_terminated_length": 594.8515625, "completions/min_length": 137.0, "completions/min_terminated_length": 137.0, "epoch": 0.3169912693082606, "frac_reward_zero_std": 0.96875, "grad_norm": 0.05815166195072682, "kl": 0.04833984375, "learning_rate": 1.4452317348132434e-06, "loss": -0.0017, "num_tokens": 223961466.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 354 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1820.0, "completions/mean_length": 580.19921875, "completions/mean_terminated_length": 574.4431762695312, "completions/min_length": 192.0, "completions/min_terminated_length": 192.0, "epoch": 0.3178867248712783, "frac_reward_zero_std": 0.9375, "grad_norm": 0.11420371586495477, "kl": 0.04827880859375, "learning_rate": 1.4333855765228104e-06, "loss": 0.0157, "num_tokens": 224564816.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "step": 355 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1698.0, "completions/mean_length": 630.931640625, "completions/mean_terminated_length": 628.1585083007812, "completions/min_length": 221.0, "completions/min_terminated_length": 221.0, "epoch": 0.31878218043429596, "frac_reward_zero_std": 1.0, "grad_norm": 0.004304656769130002, "kl": 0.0465087890625, "learning_rate": 1.421594660736675e-06, "loss": 0.0005, "num_tokens": 225199357.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 356 }, { "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": -7.0, "completions/max_length": 1453.0, "completions/max_terminated_length": 1453.0, "completions/mean_length": 544.44140625, "completions/mean_terminated_length": 544.44140625, "completions/min_length": 202.0, "completions/min_terminated_length": 202.0, "epoch": 0.31967763599731364, "frac_reward_zero_std": 1.0, "grad_norm": 0.004792128064874122, "kl": 0.05010986328125, "learning_rate": 1.4098594821780476e-06, "loss": 0.0005, "num_tokens": 225752303.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 357 }, { "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": -7.0, "completions/max_length": 1750.0, "completions/max_terminated_length": 1750.0, "completions/mean_length": 569.7265625, "completions/mean_terminated_length": 569.7265625, "completions/min_length": 206.0, "completions/min_terminated_length": 206.0, "epoch": 0.3205730915603313, "frac_reward_zero_std": 0.9375, "grad_norm": 0.07939089082655287, "kl": 0.04949951171875, "learning_rate": 1.3981805332315174e-06, "loss": 0.0001, "num_tokens": 226358003.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "step": 358 }, { "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": -7.0, "completions/max_length": 1859.0, "completions/max_terminated_length": 1859.0, "completions/mean_length": 586.755859375, "completions/mean_terminated_length": 586.755859375, "completions/min_length": 181.0, "completions/min_terminated_length": 181.0, "epoch": 0.321468547123349, "frac_reward_zero_std": 1.0, "grad_norm": 0.004747531699904538, "kl": 0.04705810546875, "learning_rate": 1.3865583039223929e-06, "loss": 0.0005, "num_tokens": 226954870.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 359 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1848.0, "completions/mean_length": 561.67578125, "completions/mean_terminated_length": 558.76708984375, "completions/min_length": 172.0, "completions/min_terminated_length": 172.0, "epoch": 0.32236400268636667, "frac_reward_zero_std": 0.96875, "grad_norm": 0.05340801461414318, "kl": 0.0465087890625, "learning_rate": 1.374993281896137e-06, "loss": 0.0166, "num_tokens": 227541488.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 360 }, { "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": -7.0, "completions/max_length": 1943.0, "completions/max_terminated_length": 1943.0, "completions/mean_length": 549.7890625, "completions/mean_terminated_length": 549.7890625, "completions/min_length": 158.0, "completions/min_terminated_length": 158.0, "epoch": 0.3232594582493844, "frac_reward_zero_std": 1.0, "grad_norm": 0.004837436430436778, "kl": 0.0484619140625, "learning_rate": 1.3634859523979134e-06, "loss": 0.0005, "num_tokens": 228105924.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 361 }, { "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": -7.0, "completions/max_length": 1845.0, "completions/max_terminated_length": 1845.0, "completions/mean_length": 572.677734375, "completions/mean_terminated_length": 572.677734375, "completions/min_length": 126.0, "completions/min_terminated_length": 126.0, "epoch": 0.3241549138124021, "frac_reward_zero_std": 1.0, "grad_norm": 0.0056382167382188035, "kl": 0.04815673828125, "learning_rate": 1.3520367982522208e-06, "loss": 0.0005, "num_tokens": 228684975.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 362 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1624.0, "completions/mean_length": 548.544921875, "completions/mean_terminated_length": 545.6105346679688, "completions/min_length": 155.0, "completions/min_terminated_length": 155.0, "epoch": 0.32505036937541976, "frac_reward_zero_std": 1.0, "grad_norm": 0.0067283143933235, "kl": 0.04949951171875, "learning_rate": 1.3406462998426358e-06, "loss": 0.0005, "num_tokens": 229248662.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 363 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1774.0, "completions/mean_length": 621.30078125, "completions/mean_terminated_length": 615.7059326171875, "completions/min_length": 235.0, "completions/min_terminated_length": 235.0, "epoch": 0.32594582493843743, "frac_reward_zero_std": 0.9375, "grad_norm": 0.08503371465958129, "kl": 0.04840087890625, "learning_rate": 1.3293149350916595e-06, "loss": 0.0103, "num_tokens": 229883920.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "step": 364 }, { "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": -7.0, "completions/max_length": 1562.0, "completions/max_terminated_length": 1562.0, "completions/mean_length": 595.625, "completions/mean_terminated_length": 595.625, "completions/min_length": 156.0, "completions/min_terminated_length": 156.0, "epoch": 0.3268412805014551, "frac_reward_zero_std": 1.0, "grad_norm": 0.0068623792592217995, "kl": 0.04730224609375, "learning_rate": 1.3180431794406623e-06, "loss": 0.0005, "num_tokens": 230488288.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 365 }, { "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": -7.0, "completions/max_length": 1877.0, "completions/max_terminated_length": 1877.0, "completions/mean_length": 582.458984375, "completions/mean_terminated_length": 582.458984375, "completions/min_length": 127.0, "completions/min_terminated_length": 127.0, "epoch": 0.3277367360644728, "frac_reward_zero_std": 1.0, "grad_norm": 0.018560148818948925, "kl": 0.04766845703125, "learning_rate": 1.3068315058299358e-06, "loss": 0.0005, "num_tokens": 231086187.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 366 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1877.0, "completions/mean_length": 613.73046875, "completions/mean_terminated_length": 608.1058959960938, "completions/min_length": 226.0, "completions/min_terminated_length": 226.0, "epoch": 0.32863219162749047, "frac_reward_zero_std": 1.0, "grad_norm": 0.006036312025660227, "kl": 0.0472412109375, "learning_rate": 1.2956803846788503e-06, "loss": 0.0005, "num_tokens": 231702705.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 367 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1561.0, "completions/mean_length": 575.67578125, "completions/mean_terminated_length": 572.7944946289062, "completions/min_length": 170.0, "completions/min_terminated_length": 170.0, "epoch": 0.32952764719050814, "frac_reward_zero_std": 1.0, "grad_norm": 0.010626533398767141, "kl": 0.04766845703125, "learning_rate": 1.284590283866116e-06, "loss": 0.0005, "num_tokens": 232314187.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 368 }, { "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": -7.0, "completions/max_length": 1877.0, "completions/max_terminated_length": 1877.0, "completions/mean_length": 638.849609375, "completions/mean_terminated_length": 638.849609375, "completions/min_length": 165.0, "completions/min_terminated_length": 165.0, "epoch": 0.3304231027535259, "frac_reward_zero_std": 1.0, "grad_norm": 0.005441410501154126, "kl": 0.04730224609375, "learning_rate": 1.2735616687101518e-06, "loss": 0.0005, "num_tokens": 233017150.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 369 }, { "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": -6.953125, "completions/max_length": 2048.0, "completions/max_terminated_length": 1908.0, "completions/mean_length": 603.595703125, "completions/mean_terminated_length": 595.08251953125, "completions/min_length": 161.0, "completions/min_terminated_length": 161.0, "epoch": 0.33131855831654355, "frac_reward_zero_std": 1.0, "grad_norm": 0.04062997694360191, "kl": 0.04681396484375, "learning_rate": 1.2625950019495614e-06, "loss": 0.0005, "num_tokens": 233633391.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 370 }, { "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": -7.0, "completions/max_length": 1448.0, "completions/max_terminated_length": 1448.0, "completions/mean_length": 572.203125, "completions/mean_terminated_length": 572.203125, "completions/min_length": 139.0, "completions/min_terminated_length": 139.0, "epoch": 0.33221401387956123, "frac_reward_zero_std": 1.0, "grad_norm": 0.00508915950543525, "kl": 0.04736328125, "learning_rate": 1.251690743723718e-06, "loss": 0.0005, "num_tokens": 234234375.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 371 }, { "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": -7.0, "completions/max_length": 1783.0, "completions/max_terminated_length": 1783.0, "completions/mean_length": 571.640625, "completions/mean_terminated_length": 571.640625, "completions/min_length": 180.0, "completions/min_terminated_length": 180.0, "epoch": 0.3331094694425789, "frac_reward_zero_std": 1.0, "grad_norm": 0.005649723189142395, "kl": 0.04681396484375, "learning_rate": 1.2408493515534581e-06, "loss": 0.0005, "num_tokens": 234813247.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 372 }, { "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": -7.0, "completions/max_length": 1921.0, "completions/max_terminated_length": 1921.0, "completions/mean_length": 582.796875, "completions/mean_terminated_length": 582.796875, "completions/min_length": 148.0, "completions/min_terminated_length": 148.0, "epoch": 0.3340049250055966, "frac_reward_zero_std": 0.96875, "grad_norm": 0.07817677753397509, "kl": 0.04779052734375, "learning_rate": 1.2300712803218834e-06, "loss": 0.0039, "num_tokens": 235400759.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 373 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1730.0, "completions/mean_length": 561.732421875, "completions/mean_terminated_length": 558.8238525390625, "completions/min_length": 117.0, "completions/min_terminated_length": 117.0, "epoch": 0.33490038056861426, "frac_reward_zero_std": 1.0, "grad_norm": 0.005750730935584701, "kl": 0.04840087890625, "learning_rate": 1.2193569822552772e-06, "loss": 0.0005, "num_tokens": 236002254.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 374 }, { "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": -7.0, "completions/max_length": 1538.0, "completions/max_terminated_length": 1538.0, "completions/mean_length": 581.84765625, "completions/mean_terminated_length": 581.84765625, "completions/min_length": 169.0, "completions/min_terminated_length": 169.0, "epoch": 0.33579583613163194, "frac_reward_zero_std": 1.0, "grad_norm": 0.005148892348158003, "kl": 0.0478515625, "learning_rate": 1.2087069069041268e-06, "loss": 0.0005, "num_tokens": 236597552.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 375 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1951.0, "completions/mean_length": 613.890625, "completions/mean_terminated_length": 608.7745361328125, "completions/min_length": 213.0, "completions/min_terminated_length": 213.0, "epoch": 0.3366912916946497, "frac_reward_zero_std": 0.96875, "grad_norm": 6.1250219766271465, "kl": 0.0712890625, "learning_rate": 1.1981215011242654e-06, "loss": 0.0117, "num_tokens": 237222104.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 376 }, { "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": -7.0, "completions/max_length": 1886.0, "completions/max_terminated_length": 1886.0, "completions/mean_length": 586.263671875, "completions/mean_terminated_length": 586.263671875, "completions/min_length": 112.0, "completions/min_terminated_length": 112.0, "epoch": 0.33758674725766735, "frac_reward_zero_std": 1.0, "grad_norm": 0.00577756812887239, "kl": 0.04656982421875, "learning_rate": 1.1876012090581184e-06, "loss": 0.0005, "num_tokens": 237836223.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 377 }, { "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": -6.984375, "completions/max_length": 1996.0, "completions/max_terminated_length": 1996.0, "completions/mean_length": 609.84765625, "completions/mean_terminated_length": 607.878662109375, "completions/min_length": 184.0, "completions/min_terminated_length": 184.0, "epoch": 0.33848220282068503, "frac_reward_zero_std": 1.0, "grad_norm": 0.02365896656045564, "kl": 0.0654296875, "learning_rate": 1.177146472116071e-06, "loss": 0.0007, "num_tokens": 238463233.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 378 }, { "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": -7.0, "completions/max_length": 1808.0, "completions/max_terminated_length": 1808.0, "completions/mean_length": 585.064453125, "completions/mean_terminated_length": 585.064453125, "completions/min_length": 153.0, "completions/min_terminated_length": 153.0, "epoch": 0.3393776583837027, "frac_reward_zero_std": 0.9375, "grad_norm": 0.07355594175930143, "kl": 0.050048828125, "learning_rate": 1.1667577289579462e-06, "loss": 0.0049, "num_tokens": 239076322.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "step": 379 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1374.0, "completions/mean_length": 589.796875, "completions/mean_terminated_length": 586.9432373046875, "completions/min_length": 197.0, "completions/min_terminated_length": 197.0, "epoch": 0.3402731139467204, "frac_reward_zero_std": 1.0, "grad_norm": 0.005923305178076053, "kl": 0.04888916015625, "learning_rate": 1.1564354154746007e-06, "loss": 0.0005, "num_tokens": 239686714.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 380 }, { "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": -7.0, "completions/max_length": 1704.0, "completions/max_terminated_length": 1704.0, "completions/mean_length": 565.86328125, "completions/mean_terminated_length": 565.86328125, "completions/min_length": 159.0, "completions/min_terminated_length": 159.0, "epoch": 0.34116856950973806, "frac_reward_zero_std": 0.96875, "grad_norm": 0.06055954764239639, "kl": 0.04974365234375, "learning_rate": 1.146179964769635e-06, "loss": 0.0006, "num_tokens": 240269876.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 381 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1780.0, "completions/mean_length": 604.6328125, "completions/mean_terminated_length": 601.8082275390625, "completions/min_length": 158.0, "completions/min_terminated_length": 158.0, "epoch": 0.34206402507275574, "frac_reward_zero_std": 0.96875, "grad_norm": 0.057122471938020274, "kl": 0.0457763671875, "learning_rate": 1.1359918071412195e-06, "loss": 0.0114, "num_tokens": 240867560.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 382 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1869.0, "completions/mean_length": 574.115234375, "completions/mean_terminated_length": 568.3353271484375, "completions/min_length": 149.0, "completions/min_terminated_length": 149.0, "epoch": 0.3429594806357735, "frac_reward_zero_std": 0.96875, "grad_norm": 0.0652873493200889, "kl": 0.04638671875, "learning_rate": 1.1258713700640456e-06, "loss": 0.014, "num_tokens": 241445155.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 383 }, { "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": -7.0, "completions/max_length": 1976.0, "completions/max_terminated_length": 1976.0, "completions/mean_length": 588.0546875, "completions/mean_terminated_length": 588.0546875, "completions/min_length": 145.0, "completions/min_terminated_length": 145.0, "epoch": 0.34385493619879115, "frac_reward_zero_std": 1.0, "grad_norm": 0.004754868273249351, "kl": 0.047607421875, "learning_rate": 1.115819078171383e-06, "loss": 0.0005, "num_tokens": 242033343.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 384 }, { "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": -7.0, "completions/max_length": 1922.0, "completions/max_terminated_length": 1922.0, "completions/mean_length": 574.859375, "completions/mean_terminated_length": 574.859375, "completions/min_length": 137.0, "completions/min_terminated_length": 137.0, "epoch": 0.3447503917618088, "frac_reward_zero_std": 1.0, "grad_norm": 0.0056589735897611725, "kl": 0.05059814453125, "learning_rate": 1.1058353532372667e-06, "loss": 0.0005, "num_tokens": 242619959.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 385 }, { "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": -6.9375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1501.0, "completions/mean_length": 573.388671875, "completions/mean_terminated_length": 563.56103515625, "completions/min_length": 151.0, "completions/min_terminated_length": 151.0, "epoch": 0.3456458473248265, "frac_reward_zero_std": 0.90625, "grad_norm": 0.11311586362691683, "kl": 0.0516357421875, "learning_rate": 1.0959206141587998e-06, "loss": 0.0303, "num_tokens": 243228654.0, "reward": 0.09941406548023224, "reward_std": 0.0023437500931322575, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.994140625, "rewards/format_reward/std": 0.07639661431312561, "step": 386 }, { "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": -7.0, "completions/max_length": 2034.0, "completions/max_terminated_length": 2034.0, "completions/mean_length": 582.330078125, "completions/mean_terminated_length": 582.330078125, "completions/min_length": 182.0, "completions/min_terminated_length": 182.0, "epoch": 0.3465413028878442, "frac_reward_zero_std": 1.0, "grad_norm": 0.005083694085424717, "kl": 0.04852294921875, "learning_rate": 1.0860752769385766e-06, "loss": 0.0005, "num_tokens": 243815575.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 387 }, { "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": -7.0, "completions/max_length": 1655.0, "completions/max_terminated_length": 1655.0, "completions/mean_length": 608.525390625, "completions/mean_terminated_length": 608.525390625, "completions/min_length": 180.0, "completions/min_terminated_length": 180.0, "epoch": 0.34743675845086186, "frac_reward_zero_std": 0.96875, "grad_norm": 0.05526627161129057, "kl": 0.0477294921875, "learning_rate": 1.0762997546672279e-06, "loss": 0.0015, "num_tokens": 244427892.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 388 }, { "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": -7.0, "completions/max_length": 1761.0, "completions/max_terminated_length": 1761.0, "completions/mean_length": 562.423828125, "completions/mean_terminated_length": 562.423828125, "completions/min_length": 149.0, "completions/min_terminated_length": 149.0, "epoch": 0.34833221401387954, "frac_reward_zero_std": 0.96875, "grad_norm": 0.08080507072024797, "kl": 0.048583984375, "learning_rate": 1.0665944575060914e-06, "loss": 0.004, "num_tokens": 245024445.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 389 }, { "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": -7.0, "completions/max_length": 1832.0, "completions/max_terminated_length": 1832.0, "completions/mean_length": 586.203125, "completions/mean_terminated_length": 586.203125, "completions/min_length": 163.0, "completions/min_terminated_length": 163.0, "epoch": 0.34922766957689727, "frac_reward_zero_std": 1.0, "grad_norm": 0.009200464557926759, "kl": 0.0494384765625, "learning_rate": 1.056959792669997e-06, "loss": 0.0005, "num_tokens": 245641717.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 390 }, { "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": -7.0, "completions/max_length": 1634.0, "completions/max_terminated_length": 1634.0, "completions/mean_length": 597.11328125, "completions/mean_terminated_length": 597.11328125, "completions/min_length": 209.0, "completions/min_terminated_length": 209.0, "epoch": 0.35012312513991495, "frac_reward_zero_std": 0.96875, "grad_norm": 0.05477299161437103, "kl": 0.04791259765625, "learning_rate": 1.0473961644101856e-06, "loss": 0.0018, "num_tokens": 246288895.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 391 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1455.0, "completions/mean_length": 610.310546875, "completions/mean_terminated_length": 607.4970703125, "completions/min_length": 184.0, "completions/min_terminated_length": 184.0, "epoch": 0.3510185807029326, "frac_reward_zero_std": 0.96875, "grad_norm": 0.04103458864719703, "kl": 0.047607421875, "learning_rate": 1.037903973997345e-06, "loss": 0.0002, "num_tokens": 246940366.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 392 }, { "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": -7.0, "completions/max_length": 1687.0, "completions/max_terminated_length": 1687.0, "completions/mean_length": 571.6171875, "completions/mean_terminated_length": 571.6171875, "completions/min_length": 165.0, "completions/min_terminated_length": 165.0, "epoch": 0.3519140362659503, "frac_reward_zero_std": 1.0, "grad_norm": 0.0045559010772451055, "kl": 0.04620361328125, "learning_rate": 1.0284836197047737e-06, "loss": 0.0005, "num_tokens": 247554154.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 393 }, { "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": -7.0, "completions/max_length": 1482.0, "completions/max_terminated_length": 1482.0, "completions/mean_length": 529.73046875, "completions/mean_terminated_length": 529.73046875, "completions/min_length": 171.0, "completions/min_terminated_length": 171.0, "epoch": 0.352809491828968, "frac_reward_zero_std": 1.0, "grad_norm": 0.005620991916216765, "kl": 0.05078125, "learning_rate": 1.0191354967916712e-06, "loss": 0.0005, "num_tokens": 248125712.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 394 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1438.0, "completions/mean_length": 577.833984375, "completions/mean_terminated_length": 574.9569702148438, "completions/min_length": 151.0, "completions/min_terminated_length": 151.0, "epoch": 0.35370494739198566, "frac_reward_zero_std": 1.0, "grad_norm": 0.005897057155196266, "kl": 0.05084228515625, "learning_rate": 1.0098599974865515e-06, "loss": 0.0005, "num_tokens": 248716379.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 395 }, { "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": -7.0, "completions/max_length": 1838.0, "completions/max_terminated_length": 1838.0, "completions/mean_length": 607.990234375, "completions/mean_terminated_length": 607.990234375, "completions/min_length": 156.0, "completions/min_terminated_length": 156.0, "epoch": 0.35460040295500334, "frac_reward_zero_std": 1.0, "grad_norm": 0.005227169861081531, "kl": 0.04913330078125, "learning_rate": 1.0006575109707898e-06, "loss": 0.0005, "num_tokens": 249353110.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 396 }, { "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": -7.0, "completions/max_length": 1525.0, "completions/max_terminated_length": 1525.0, "completions/mean_length": 534.1484375, "completions/mean_terminated_length": 534.1484375, "completions/min_length": 97.0, "completions/min_terminated_length": 97.0, "epoch": 0.35549585851802107, "frac_reward_zero_std": 1.0, "grad_norm": 0.009091379046012489, "kl": 0.04925537109375, "learning_rate": 9.915284233622877e-07, "loss": 0.0005, "num_tokens": 249943106.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 397 }, { "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": -7.0, "completions/max_length": 1687.0, "completions/max_terminated_length": 1687.0, "completions/mean_length": 554.1640625, "completions/mean_terminated_length": 554.1640625, "completions/min_length": 110.0, "completions/min_terminated_length": 110.0, "epoch": 0.35639131408103875, "frac_reward_zero_std": 1.0, "grad_norm": 0.00576347525077593, "kl": 0.04864501953125, "learning_rate": 9.824731176992796e-07, "loss": 0.0005, "num_tokens": 250492534.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 398 }, { "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": -7.0, "completions/max_length": 1398.0, "completions/max_terminated_length": 1398.0, "completions/mean_length": 580.962890625, "completions/mean_terminated_length": 580.962890625, "completions/min_length": 166.0, "completions/min_terminated_length": 166.0, "epoch": 0.3572867696440564, "frac_reward_zero_std": 1.0, "grad_norm": 0.005684393564410186, "kl": 0.05072021484375, "learning_rate": 9.734919739242543e-07, "loss": 0.0005, "num_tokens": 251093283.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 399 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1882.0, "completions/mean_length": 578.701171875, "completions/mean_terminated_length": 572.9392700195312, "completions/min_length": 156.0, "completions/min_terminated_length": 156.0, "epoch": 0.3581822252070741, "frac_reward_zero_std": 0.9375, "grad_norm": 0.09024918016321634, "kl": 0.04998779296875, "learning_rate": 9.645853688680177e-07, "loss": 0.0057, "num_tokens": 251716346.0, "reward": 0.09941406548023224, "reward_std": 0.0018486406188458204, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.994140625, "rewards/format_reward/std": 0.07639661431312561, "step": 400 }, { "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": -7.0, "completions/max_length": 1275.0, "completions/max_terminated_length": 1275.0, "completions/mean_length": 533.146484375, "completions/mean_terminated_length": 533.146484375, "completions/min_length": 121.0, "completions/min_terminated_length": 121.0, "epoch": 0.3590776807700918, "frac_reward_zero_std": 0.96875, "grad_norm": 0.07662408160218846, "kl": 0.04638671875, "learning_rate": 9.557536762338786e-07, "loss": 0.0079, "num_tokens": 252276213.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 401 }, { "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": -7.0, "completions/max_length": 1603.0, "completions/max_terminated_length": 1603.0, "completions/mean_length": 539.779296875, "completions/mean_terminated_length": 539.779296875, "completions/min_length": 138.0, "completions/min_terminated_length": 138.0, "epoch": 0.35997313633310946, "frac_reward_zero_std": 1.0, "grad_norm": 0.0047643547604445105, "kl": 0.04693603515625, "learning_rate": 9.46997266581973e-07, "loss": 0.0005, "num_tokens": 252837172.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 402 }, { "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": -7.0, "completions/max_length": 1944.0, "completions/max_terminated_length": 1944.0, "completions/mean_length": 597.201171875, "completions/mean_terminated_length": 597.201171875, "completions/min_length": 171.0, "completions/min_terminated_length": 171.0, "epoch": 0.36086859189612713, "frac_reward_zero_std": 1.0, "grad_norm": 0.004834734143787087, "kl": 0.0479736328125, "learning_rate": 9.383165073137115e-07, "loss": 0.0005, "num_tokens": 253443979.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 403 }, { "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": -7.0, "completions/max_length": 1626.0, "completions/max_terminated_length": 1626.0, "completions/mean_length": 566.322265625, "completions/mean_terminated_length": 566.322265625, "completions/min_length": 165.0, "completions/min_terminated_length": 165.0, "epoch": 0.36176404745914487, "frac_reward_zero_std": 1.0, "grad_norm": 0.00579926871259698, "kl": 0.04705810546875, "learning_rate": 9.297117626563687e-07, "loss": 0.0005, "num_tokens": 254029248.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 404 }, { "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": -7.0, "completions/max_length": 1328.0, "completions/max_terminated_length": 1328.0, "completions/mean_length": 586.23046875, "completions/mean_terminated_length": 586.23046875, "completions/min_length": 207.0, "completions/min_terminated_length": 207.0, "epoch": 0.36265950302216254, "frac_reward_zero_std": 1.0, "grad_norm": 0.005531919514784843, "kl": 0.048583984375, "learning_rate": 9.211833936477957e-07, "loss": 0.0005, "num_tokens": 254609734.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 405 }, { "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": -7.0, "completions/max_length": 1538.0, "completions/max_terminated_length": 1538.0, "completions/mean_length": 562.662109375, "completions/mean_terminated_length": 562.662109375, "completions/min_length": 164.0, "completions/min_terminated_length": 164.0, "epoch": 0.3635549585851802, "frac_reward_zero_std": 1.0, "grad_norm": 0.00421872727224393, "kl": 0.04925537109375, "learning_rate": 9.127317581212753e-07, "loss": 0.0005, "num_tokens": 255193529.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 406 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1312.0, "completions/mean_length": 546.47265625, "completions/mean_terminated_length": 542.1961059570312, "completions/min_length": 175.0, "completions/min_terminated_length": 175.0, "epoch": 0.3644504141481979, "frac_reward_zero_std": 0.96875, "grad_norm": 0.10098327848008554, "kl": 0.05133056640625, "learning_rate": 9.043572106905084e-07, "loss": 0.013, "num_tokens": 255796619.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 407 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1818.0, "completions/mean_length": 574.74609375, "completions/mean_terminated_length": 571.863037109375, "completions/min_length": 194.0, "completions/min_terminated_length": 194.0, "epoch": 0.3653458697112156, "frac_reward_zero_std": 0.96875, "grad_norm": 0.07221448169919696, "kl": 0.04937744140625, "learning_rate": 8.960601027347321e-07, "loss": 0.0011, "num_tokens": 256396969.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 408 }, { "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": -6.984375, "completions/max_length": 1660.0, "completions/max_terminated_length": 1660.0, "completions/mean_length": 531.146484375, "completions/mean_terminated_length": 530.3033447265625, "completions/min_length": 162.0, "completions/min_terminated_length": 162.0, "epoch": 0.36624132527423325, "frac_reward_zero_std": 0.96875, "grad_norm": 0.06674911247595358, "kl": 0.04931640625, "learning_rate": 8.878407823839788e-07, "loss": 0.0029, "num_tokens": 256954724.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 409 }, { "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": -6.984375, "completions/max_length": 2013.0, "completions/max_terminated_length": 2013.0, "completions/mean_length": 568.6484375, "completions/mean_terminated_length": 567.4246826171875, "completions/min_length": 136.0, "completions/min_terminated_length": 136.0, "epoch": 0.36713678083725093, "frac_reward_zero_std": 1.0, "grad_norm": 0.010132848557988017, "kl": 0.05450439453125, "learning_rate": 8.796995945044689e-07, "loss": 0.0005, "num_tokens": 257552176.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 410 }, { "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": -7.0, "completions/max_length": 1686.0, "completions/max_terminated_length": 1686.0, "completions/mean_length": 581.15234375, "completions/mean_terminated_length": 581.15234375, "completions/min_length": 159.0, "completions/min_terminated_length": 159.0, "epoch": 0.36803223640026866, "frac_reward_zero_std": 1.0, "grad_norm": 0.00432008518918132, "kl": 0.0484619140625, "learning_rate": 8.716368806841405e-07, "loss": 0.0005, "num_tokens": 258158942.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 411 }, { "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": -7.0, "completions/max_length": 1425.0, "completions/max_terminated_length": 1425.0, "completions/mean_length": 561.341796875, "completions/mean_terminated_length": 561.341796875, "completions/min_length": 162.0, "completions/min_terminated_length": 162.0, "epoch": 0.36892769196328634, "frac_reward_zero_std": 0.96875, "grad_norm": 0.06693413618462407, "kl": 0.0494384765625, "learning_rate": 8.636529792183171e-07, "loss": -0.001, "num_tokens": 258739869.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 412 }, { "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": -7.0, "completions/max_length": 1556.0, "completions/max_terminated_length": 1556.0, "completions/mean_length": 560.208984375, "completions/mean_terminated_length": 560.208984375, "completions/min_length": 152.0, "completions/min_terminated_length": 152.0, "epoch": 0.369823147526304, "frac_reward_zero_std": 1.0, "grad_norm": 0.004847298445441672, "kl": 0.04901123046875, "learning_rate": 8.557482250955144e-07, "loss": 0.0005, "num_tokens": 259334584.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 413 }, { "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": -7.0, "completions/max_length": 1558.0, "completions/max_terminated_length": 1558.0, "completions/mean_length": 611.892578125, "completions/mean_terminated_length": 611.892578125, "completions/min_length": 198.0, "completions/min_terminated_length": 198.0, "epoch": 0.3707186030893217, "frac_reward_zero_std": 1.0, "grad_norm": 0.005198359727407881, "kl": 0.04888916015625, "learning_rate": 8.479229499833844e-07, "loss": 0.0005, "num_tokens": 259972241.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 414 }, { "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": -7.0, "completions/max_length": 1332.0, "completions/max_terminated_length": 1332.0, "completions/mean_length": 529.185546875, "completions/mean_terminated_length": 529.185546875, "completions/min_length": 135.0, "completions/min_terminated_length": 135.0, "epoch": 0.3716140586523394, "frac_reward_zero_std": 1.0, "grad_norm": 0.005096616784905531, "kl": 0.04815673828125, "learning_rate": 8.401774822147976e-07, "loss": 0.0005, "num_tokens": 260477904.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 415 }, { "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": -7.0, "completions/max_length": 1589.0, "completions/max_terminated_length": 1589.0, "completions/mean_length": 574.904296875, "completions/mean_terminated_length": 574.904296875, "completions/min_length": 155.0, "completions/min_terminated_length": 155.0, "epoch": 0.37250951421535705, "frac_reward_zero_std": 0.96875, "grad_norm": 0.06747598747854207, "kl": 0.04888916015625, "learning_rate": 8.325121467740695e-07, "loss": 0.001, "num_tokens": 261093247.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 416 }, { "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": -7.0, "completions/max_length": 1710.0, "completions/max_terminated_length": 1710.0, "completions/mean_length": 573.78515625, "completions/mean_terminated_length": 573.78515625, "completions/min_length": 122.0, "completions/min_terminated_length": 122.0, "epoch": 0.37340496977837473, "frac_reward_zero_std": 1.0, "grad_norm": 0.004344603358267009, "kl": 0.0465087890625, "learning_rate": 8.249272652833226e-07, "loss": 0.0005, "num_tokens": 261713041.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 417 }, { "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": -7.0, "completions/max_length": 1463.0, "completions/max_terminated_length": 1463.0, "completions/mean_length": 533.013671875, "completions/mean_terminated_length": 533.013671875, "completions/min_length": 175.0, "completions/min_terminated_length": 175.0, "epoch": 0.3743004253413924, "frac_reward_zero_std": 1.0, "grad_norm": 0.004507059752267276, "kl": 0.0506591796875, "learning_rate": 8.174231559889931e-07, "loss": 0.0005, "num_tokens": 262281848.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 418 }, { "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": -7.0, "completions/max_length": 1396.0, "completions/max_terminated_length": 1396.0, "completions/mean_length": 517.865234375, "completions/mean_terminated_length": 517.865234375, "completions/min_length": 136.0, "completions/min_terminated_length": 136.0, "epoch": 0.37519588090441014, "frac_reward_zero_std": 1.0, "grad_norm": 0.005351908786787514, "kl": 0.04852294921875, "learning_rate": 8.100001337484787e-07, "loss": 0.0005, "num_tokens": 262811091.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 419 }, { "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": -7.0, "completions/max_length": 1760.0, "completions/max_terminated_length": 1760.0, "completions/mean_length": 561.669921875, "completions/mean_terminated_length": 561.669921875, "completions/min_length": 185.0, "completions/min_terminated_length": 185.0, "epoch": 0.3760913364674278, "frac_reward_zero_std": 0.96875, "grad_norm": 0.058773060593634845, "kl": 0.0477294921875, "learning_rate": 8.026585100169251e-07, "loss": -0.0004, "num_tokens": 263420570.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 420 }, { "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": -7.0, "completions/max_length": 1659.0, "completions/max_terminated_length": 1659.0, "completions/mean_length": 601.13671875, "completions/mean_terminated_length": 601.13671875, "completions/min_length": 146.0, "completions/min_terminated_length": 146.0, "epoch": 0.3769867920304455, "frac_reward_zero_std": 0.96875, "grad_norm": 0.04780470323490276, "kl": 0.04840087890625, "learning_rate": 7.953985928341601e-07, "loss": -0.0011, "num_tokens": 264049296.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 421 }, { "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": -7.0, "completions/max_length": 1901.0, "completions/max_terminated_length": 1901.0, "completions/mean_length": 587.703125, "completions/mean_terminated_length": 587.703125, "completions/min_length": 122.0, "completions/min_terminated_length": 122.0, "epoch": 0.37788224759346317, "frac_reward_zero_std": 1.0, "grad_norm": 0.015584951802210437, "kl": 0.0504150390625, "learning_rate": 7.882206868117693e-07, "loss": 0.0005, "num_tokens": 264649528.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 422 }, { "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": -7.0, "completions/max_length": 1931.0, "completions/max_terminated_length": 1931.0, "completions/mean_length": 575.669921875, "completions/mean_terminated_length": 575.669921875, "completions/min_length": 145.0, "completions/min_terminated_length": 145.0, "epoch": 0.37877770315648085, "frac_reward_zero_std": 1.0, "grad_norm": 0.008202181121387101, "kl": 0.04742431640625, "learning_rate": 7.81125093120313e-07, "loss": 0.0005, "num_tokens": 265265791.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 423 }, { "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": -7.0, "completions/max_length": 1773.0, "completions/max_terminated_length": 1773.0, "completions/mean_length": 554.826171875, "completions/mean_terminated_length": 554.826171875, "completions/min_length": 153.0, "completions/min_terminated_length": 153.0, "epoch": 0.3796731587194985, "frac_reward_zero_std": 0.96875, "grad_norm": 0.07411872102018555, "kl": 0.04827880859375, "learning_rate": 7.741121094766916e-07, "loss": 0.0036, "num_tokens": 265876198.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 424 }, { "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": -7.0, "completions/max_length": 1806.0, "completions/max_terminated_length": 1806.0, "completions/mean_length": 531.36328125, "completions/mean_terminated_length": 531.36328125, "completions/min_length": 116.0, "completions/min_terminated_length": 116.0, "epoch": 0.3805686142825162, "frac_reward_zero_std": 1.0, "grad_norm": 0.00483447721173187, "kl": 0.049072265625, "learning_rate": 7.671820301316532e-07, "loss": 0.0005, "num_tokens": 266424928.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 425 }, { "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": -7.0, "completions/max_length": 1398.0, "completions/max_terminated_length": 1398.0, "completions/mean_length": 547.166015625, "completions/mean_terminated_length": 547.166015625, "completions/min_length": 161.0, "completions/min_terminated_length": 161.0, "epoch": 0.38146406984553394, "frac_reward_zero_std": 1.0, "grad_norm": 0.004354827538356667, "kl": 0.04656982421875, "learning_rate": 7.603351458574474e-07, "loss": 0.0005, "num_tokens": 266996373.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 426 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1578.0, "completions/mean_length": 560.0390625, "completions/mean_terminated_length": 557.127197265625, "completions/min_length": 162.0, "completions/min_terminated_length": 162.0, "epoch": 0.3823595254085516, "frac_reward_zero_std": 0.96875, "grad_norm": 0.052963479074057004, "kl": 0.05413818359375, "learning_rate": 7.535717439356255e-07, "loss": 0.0009, "num_tokens": 267662393.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 427 }, { "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": -7.0, "completions/max_length": 1869.0, "completions/max_terminated_length": 1869.0, "completions/mean_length": 532.46484375, "completions/mean_terminated_length": 532.46484375, "completions/min_length": 199.0, "completions/min_terminated_length": 199.0, "epoch": 0.3832549809715693, "frac_reward_zero_std": 1.0, "grad_norm": 0.14959490402999703, "kl": 0.0740966796875, "learning_rate": 7.46892108144986e-07, "loss": 0.0007, "num_tokens": 268222999.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 428 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1312.0, "completions/mean_length": 572.41015625, "completions/mean_terminated_length": 569.5225219726562, "completions/min_length": 181.0, "completions/min_terminated_length": 181.0, "epoch": 0.38415043653458697, "frac_reward_zero_std": 1.0, "grad_norm": 0.005582201559752547, "kl": 0.05035400390625, "learning_rate": 7.402965187496697e-07, "loss": 0.0005, "num_tokens": 268835369.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 429 }, { "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": -7.0, "completions/max_length": 1578.0, "completions/max_terminated_length": 1578.0, "completions/mean_length": 572.984375, "completions/mean_terminated_length": 572.984375, "completions/min_length": 197.0, "completions/min_terminated_length": 197.0, "epoch": 0.38504589209760465, "frac_reward_zero_std": 0.96875, "grad_norm": 0.05576338510401166, "kl": 0.0494384765625, "learning_rate": 7.337852524873974e-07, "loss": 0.0035, "num_tokens": 269440033.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 430 }, { "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": -7.0, "completions/max_length": 1423.0, "completions/max_terminated_length": 1423.0, "completions/mean_length": 562.912109375, "completions/mean_terminated_length": 562.912109375, "completions/min_length": 170.0, "completions/min_terminated_length": 170.0, "epoch": 0.3859413476606223, "frac_reward_zero_std": 0.96875, "grad_norm": 0.08650265338397081, "kl": 0.04925537109375, "learning_rate": 7.273585825578608e-07, "loss": 0.0027, "num_tokens": 269999428.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 431 }, { "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": -7.0, "completions/max_length": 1740.0, "completions/max_terminated_length": 1740.0, "completions/mean_length": 583.6328125, "completions/mean_terminated_length": 583.6328125, "completions/min_length": 204.0, "completions/min_terminated_length": 204.0, "epoch": 0.38683680322364, "frac_reward_zero_std": 1.0, "grad_norm": 0.0041924114517845655, "kl": 0.04949951171875, "learning_rate": 7.21016778611259e-07, "loss": 0.0005, "num_tokens": 270609576.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 432 }, { "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": -7.0, "completions/max_length": 1501.0, "completions/max_terminated_length": 1501.0, "completions/mean_length": 555.6171875, "completions/mean_terminated_length": 555.6171875, "completions/min_length": 119.0, "completions/min_terminated_length": 119.0, "epoch": 0.38773225878665774, "frac_reward_zero_std": 1.0, "grad_norm": 0.004051015296435293, "kl": 0.04632568359375, "learning_rate": 7.147601067369835e-07, "loss": 0.0005, "num_tokens": 271192932.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 433 }, { "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": -7.0, "completions/max_length": 1703.0, "completions/max_terminated_length": 1703.0, "completions/mean_length": 600.853515625, "completions/mean_terminated_length": 600.853515625, "completions/min_length": 199.0, "completions/min_terminated_length": 199.0, "epoch": 0.3886277143496754, "frac_reward_zero_std": 1.0, "grad_norm": 0.003873089336388401, "kl": 0.047607421875, "learning_rate": 7.085888294524561e-07, "loss": 0.0005, "num_tokens": 271807193.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 434 }, { "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": -7.0, "completions/max_length": 1880.0, "completions/max_terminated_length": 1880.0, "completions/mean_length": 589.169921875, "completions/mean_terminated_length": 589.169921875, "completions/min_length": 130.0, "completions/min_terminated_length": 130.0, "epoch": 0.3895231699126931, "frac_reward_zero_std": 0.96875, "grad_norm": 0.06598911812763071, "kl": 0.04876708984375, "learning_rate": 7.025032056921117e-07, "loss": 0.0017, "num_tokens": 272413184.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 435 }, { "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": -7.0, "completions/max_length": 1621.0, "completions/max_terminated_length": 1621.0, "completions/mean_length": 554.048828125, "completions/mean_terminated_length": 554.048828125, "completions/min_length": 176.0, "completions/min_terminated_length": 176.0, "epoch": 0.39041862547571077, "frac_reward_zero_std": 1.0, "grad_norm": 0.005011629822964105, "kl": 0.04913330078125, "learning_rate": 6.965034907965349e-07, "loss": 0.0005, "num_tokens": 273018809.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 436 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1509.0, "completions/mean_length": 546.236328125, "completions/mean_terminated_length": 540.3471069335938, "completions/min_length": 138.0, "completions/min_terminated_length": 138.0, "epoch": 0.39131408103872845, "frac_reward_zero_std": 0.96875, "grad_norm": 0.05460977307454807, "kl": 0.0478515625, "learning_rate": 6.905899365017462e-07, "loss": 0.0158, "num_tokens": 273584962.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 437 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1624.0, "completions/mean_length": 561.583984375, "completions/mean_terminated_length": 558.6751708984375, "completions/min_length": 157.0, "completions/min_terminated_length": 157.0, "epoch": 0.3922095366017461, "frac_reward_zero_std": 1.0, "grad_norm": 0.0062424353732860646, "kl": 0.05084228515625, "learning_rate": 6.847627909286409e-07, "loss": 0.0005, "num_tokens": 274214125.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 438 }, { "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": -7.0, "completions/max_length": 1474.0, "completions/max_terminated_length": 1474.0, "completions/mean_length": 580.083984375, "completions/mean_terminated_length": 580.083984375, "completions/min_length": 133.0, "completions/min_terminated_length": 133.0, "epoch": 0.3931049921647638, "frac_reward_zero_std": 1.0, "grad_norm": 0.007684600815592728, "kl": 0.0478515625, "learning_rate": 6.790222985725761e-07, "loss": 0.0005, "num_tokens": 274811064.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 439 }, { "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": -6.75, "completions/max_length": 1687.0, "completions/max_terminated_length": 1687.0, "completions/mean_length": 562.736328125, "completions/mean_terminated_length": 551.4375, "completions/min_length": 189.0, "completions/min_terminated_length": 189.0, "epoch": 0.39400044772778153, "frac_reward_zero_std": 1.0, "grad_norm": 105.4794447937914, "kl": 12.5390625, "learning_rate": 6.733687002931141e-07, "loss": 0.1253, "num_tokens": 275402609.0, "reward": 0.09687499701976776, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.96875, "rewards/format_reward/std": 0.17416280508041382, "step": 440 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1720.0, "completions/mean_length": 574.7421875, "completions/mean_terminated_length": 568.9647216796875, "completions/min_length": 154.0, "completions/min_terminated_length": 154.0, "epoch": 0.3948959032907992, "frac_reward_zero_std": 0.875, "grad_norm": 0.15188148089603382, "kl": 0.0477294921875, "learning_rate": 6.678022333039158e-07, "loss": 0.024, "num_tokens": 275982381.0, "reward": 0.09921875596046448, "reward_std": 0.0031250000465661287, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.9921875, "rewards/format_reward/std": 0.08812850713729858, "step": 441 }, { "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": -7.0, "completions/max_length": 1923.0, "completions/max_terminated_length": 1923.0, "completions/mean_length": 606.69921875, "completions/mean_terminated_length": 606.69921875, "completions/min_length": 161.0, "completions/min_terminated_length": 161.0, "epoch": 0.3957913588538169, "frac_reward_zero_std": 0.96875, "grad_norm": 0.06622735966778687, "kl": 0.04669189453125, "learning_rate": 6.623231311627876e-07, "loss": 0.0043, "num_tokens": 276612803.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 442 }, { "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": -7.0, "completions/max_length": 1895.0, "completions/max_terminated_length": 1895.0, "completions/mean_length": 589.787109375, "completions/mean_terminated_length": 589.787109375, "completions/min_length": 163.0, "completions/min_terminated_length": 163.0, "epoch": 0.39668681441683457, "frac_reward_zero_std": 1.0, "grad_norm": 0.004054937917658486, "kl": 0.04730224609375, "learning_rate": 6.569316237618811e-07, "loss": 0.0005, "num_tokens": 277203622.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 443 }, { "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": -6.96875, "completions/max_length": 1854.0, "completions/max_terminated_length": 1660.0, "completions/mean_length": 602.287109375, "completions/mean_terminated_length": 597.3784790039062, "completions/min_length": 164.0, "completions/min_terminated_length": 164.0, "epoch": 0.39758226997985224, "frac_reward_zero_std": 0.96875, "grad_norm": 0.07219201096985092, "kl": 0.04864501953125, "learning_rate": 6.516279373180499e-07, "loss": 0.0084, "num_tokens": 277847913.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 444 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1784.0, "completions/mean_length": 554.794921875, "completions/mean_terminated_length": 551.872802734375, "completions/min_length": 138.0, "completions/min_terminated_length": 138.0, "epoch": 0.3984777255428699, "frac_reward_zero_std": 1.0, "grad_norm": 0.004333441625319936, "kl": 0.0489501953125, "learning_rate": 6.464122943633543e-07, "loss": 0.0005, "num_tokens": 278461056.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 445 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1661.0, "completions/mean_length": 592.8828125, "completions/mean_terminated_length": 590.0352172851562, "completions/min_length": 174.0, "completions/min_terminated_length": 174.0, "epoch": 0.3993731811058876, "frac_reward_zero_std": 0.96875, "grad_norm": 0.07587832755649311, "kl": 0.0469970703125, "learning_rate": 6.412849137357271e-07, "loss": -0.0016, "num_tokens": 279083252.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 446 }, { "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": -6.984375, "completions/max_length": 1835.0, "completions/max_terminated_length": 1835.0, "completions/mean_length": 558.845703125, "completions/mean_terminated_length": 558.0822143554688, "completions/min_length": 147.0, "completions/min_terminated_length": 147.0, "epoch": 0.40026863666890533, "frac_reward_zero_std": 0.9375, "grad_norm": 5.034003120090257, "kl": 0.0579833984375, "learning_rate": 6.3624601056979e-07, "loss": 0.0124, "num_tokens": 279681589.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "step": 447 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1953.0, "completions/mean_length": 614.783203125, "completions/mean_terminated_length": 611.9784545898438, "completions/min_length": 248.0, "completions/min_terminated_length": 248.0, "epoch": 0.401164092231923, "frac_reward_zero_std": 0.96875, "grad_norm": 0.04124554822003232, "kl": 0.048583984375, "learning_rate": 6.312957962878278e-07, "loss": -0.0048, "num_tokens": 280269110.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 448 }, { "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": -7.0, "completions/max_length": 1602.0, "completions/max_terminated_length": 1602.0, "completions/mean_length": 570.455078125, "completions/mean_terminated_length": 570.455078125, "completions/min_length": 169.0, "completions/min_terminated_length": 169.0, "epoch": 0.4020595477949407, "frac_reward_zero_std": 1.0, "grad_norm": 0.004252061047808049, "kl": 0.0467529296875, "learning_rate": 6.264344785909181e-07, "loss": 0.0005, "num_tokens": 280854959.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 449 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1850.0, "completions/mean_length": 615.158203125, "completions/mean_terminated_length": 610.9019775390625, "completions/min_length": 225.0, "completions/min_terminated_length": 225.0, "epoch": 0.40295500335795836, "frac_reward_zero_std": 0.96875, "grad_norm": 4.155482563277789, "kl": 0.0538330078125, "learning_rate": 6.216622614502149e-07, "loss": 0.0073, "num_tokens": 281508256.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 450 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1883.0, "completions/mean_length": 618.771484375, "completions/mean_terminated_length": 615.9745483398438, "completions/min_length": 186.0, "completions/min_terminated_length": 186.0, "epoch": 0.40385045892097604, "frac_reward_zero_std": 0.9375, "grad_norm": 0.069402280555639, "kl": 0.0484619140625, "learning_rate": 6.169793450983916e-07, "loss": -0.0053, "num_tokens": 282139451.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "step": 451 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1601.0, "completions/mean_length": 629.6171875, "completions/mean_terminated_length": 624.054931640625, "completions/min_length": 181.0, "completions/min_terminated_length": 181.0, "epoch": 0.4047459144839937, "frac_reward_zero_std": 0.96875, "grad_norm": 0.05619296797430902, "kl": 0.04974365234375, "learning_rate": 6.123859260212393e-07, "loss": -0.0006, "num_tokens": 282805479.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 452 }, { "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": -7.0, "completions/max_length": 1345.0, "completions/max_terminated_length": 1345.0, "completions/mean_length": 551.099609375, "completions/mean_terminated_length": 551.099609375, "completions/min_length": 164.0, "completions/min_terminated_length": 164.0, "epoch": 0.4056413700470114, "frac_reward_zero_std": 0.96875, "grad_norm": 0.06716769755390013, "kl": 0.047607421875, "learning_rate": 6.07882196949423e-07, "loss": 0.0002, "num_tokens": 283388314.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 453 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1887.0, "completions/mean_length": 585.65625, "completions/mean_terminated_length": 582.7944946289062, "completions/min_length": 209.0, "completions/min_terminated_length": 209.0, "epoch": 0.40653682561002913, "frac_reward_zero_std": 1.0, "grad_norm": 0.004697335407404541, "kl": 0.04742431640625, "learning_rate": 6.034683468503948e-07, "loss": 0.0005, "num_tokens": 283968202.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 454 }, { "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": -6.75, "completions/max_length": 1486.0, "completions/max_terminated_length": 1486.0, "completions/mean_length": 585.796875, "completions/mean_terminated_length": 574.6612548828125, "completions/min_length": 163.0, "completions/min_terminated_length": 163.0, "epoch": 0.4074322811730468, "frac_reward_zero_std": 1.0, "grad_norm": 1.027380515611281, "kl": 0.306640625, "learning_rate": 5.991445609204641e-07, "loss": 0.0031, "num_tokens": 284611314.0, "reward": 0.09687499701976776, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.96875, "rewards/format_reward/std": 0.17416280508041382, "step": 455 }, { "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": -7.0, "completions/max_length": 1545.0, "completions/max_terminated_length": 1545.0, "completions/mean_length": 546.5234375, "completions/mean_terminated_length": 546.5234375, "completions/min_length": 149.0, "completions/min_terminated_length": 149.0, "epoch": 0.4083277367360645, "frac_reward_zero_std": 0.96875, "grad_norm": 0.0556622348002105, "kl": 0.04693603515625, "learning_rate": 5.949110205770292e-07, "loss": -0.0011, "num_tokens": 285224430.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 456 }, { "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": -7.0, "completions/max_length": 1539.0, "completions/max_terminated_length": 1539.0, "completions/mean_length": 569.07421875, "completions/mean_terminated_length": 569.07421875, "completions/min_length": 179.0, "completions/min_terminated_length": 179.0, "epoch": 0.40922319229908216, "frac_reward_zero_std": 0.96875, "grad_norm": 0.038169107742384197, "kl": 0.0482177734375, "learning_rate": 5.90767903450964e-07, "loss": -0.0023, "num_tokens": 285823204.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 457 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1509.0, "completions/mean_length": 579.447265625, "completions/mean_terminated_length": 576.5733642578125, "completions/min_length": 188.0, "completions/min_terminated_length": 188.0, "epoch": 0.41011864786209984, "frac_reward_zero_std": 1.0, "grad_norm": 0.005065842696717293, "kl": 0.047607421875, "learning_rate": 5.867153833791652e-07, "loss": 0.0005, "num_tokens": 286402665.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 458 }, { "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": -6.953125, "completions/max_length": 2048.0, "completions/max_terminated_length": 1992.0, "completions/mean_length": 614.25, "completions/mean_terminated_length": 605.7996215820312, "completions/min_length": 112.0, "completions/min_terminated_length": 112.0, "epoch": 0.4110141034251175, "frac_reward_zero_std": 1.0, "grad_norm": 0.003912700365631719, "kl": 0.04583740234375, "learning_rate": 5.827536303972587e-07, "loss": 0.0005, "num_tokens": 286987241.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 459 }, { "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": -7.0, "completions/max_length": 1813.0, "completions/max_terminated_length": 1813.0, "completions/mean_length": 560.013671875, "completions/mean_terminated_length": 560.013671875, "completions/min_length": 69.0, "completions/min_terminated_length": 69.0, "epoch": 0.4119095589881352, "frac_reward_zero_std": 1.0, "grad_norm": 0.004495489669199859, "kl": 0.04718017578125, "learning_rate": 5.78882810732465e-07, "loss": 0.0005, "num_tokens": 287555536.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 460 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1580.0, "completions/mean_length": 558.630859375, "completions/mean_terminated_length": 555.7162475585938, "completions/min_length": 130.0, "completions/min_terminated_length": 130.0, "epoch": 0.4128050145511529, "frac_reward_zero_std": 1.0, "grad_norm": 0.0046506916985020715, "kl": 0.0484619140625, "learning_rate": 5.75103086796625e-07, "loss": 0.0005, "num_tokens": 288162883.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 461 }, { "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": -7.0, "completions/max_length": 1877.0, "completions/max_terminated_length": 1877.0, "completions/mean_length": 624.595703125, "completions/mean_terminated_length": 624.595703125, "completions/min_length": 239.0, "completions/min_terminated_length": 239.0, "epoch": 0.4137004701141706, "frac_reward_zero_std": 1.0, "grad_norm": 0.0049626766542558935, "kl": 0.0458984375, "learning_rate": 5.714146171793846e-07, "loss": 0.0005, "num_tokens": 288826548.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 462 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1807.0, "completions/mean_length": 675.462890625, "completions/mean_terminated_length": 670.0804443359375, "completions/min_length": 169.0, "completions/min_terminated_length": 169.0, "epoch": 0.4145959256771883, "frac_reward_zero_std": 1.0, "grad_norm": 0.008119931529559445, "kl": 0.049072265625, "learning_rate": 5.678175566415422e-07, "loss": 0.0005, "num_tokens": 289538001.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 463 }, { "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": -7.0, "completions/max_length": 1913.0, "completions/max_terminated_length": 1913.0, "completions/mean_length": 582.626953125, "completions/mean_terminated_length": 582.626953125, "completions/min_length": 129.0, "completions/min_terminated_length": 129.0, "epoch": 0.41549138124020596, "frac_reward_zero_std": 1.0, "grad_norm": 0.004928378991447442, "kl": 0.049072265625, "learning_rate": 5.643120561085528e-07, "loss": 0.0005, "num_tokens": 290164018.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 464 }, { "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": -7.0, "completions/max_length": 1680.0, "completions/max_terminated_length": 1680.0, "completions/mean_length": 601.583984375, "completions/mean_terminated_length": 601.583984375, "completions/min_length": 217.0, "completions/min_terminated_length": 217.0, "epoch": 0.41638683680322364, "frac_reward_zero_std": 1.0, "grad_norm": 0.008088857980013029, "kl": 0.04791259765625, "learning_rate": 5.608982626641991e-07, "loss": 0.0005, "num_tokens": 290800349.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 465 }, { "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": -7.0, "completions/max_length": 1930.0, "completions/max_terminated_length": 1930.0, "completions/mean_length": 568.7578125, "completions/mean_terminated_length": 568.7578125, "completions/min_length": 81.0, "completions/min_terminated_length": 81.0, "epoch": 0.4172822923662413, "frac_reward_zero_std": 1.0, "grad_norm": 0.024017793326594178, "kl": 0.05133056640625, "learning_rate": 5.575763195444166e-07, "loss": 0.0005, "num_tokens": 291339153.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 466 }, { "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": -7.0, "completions/max_length": 1584.0, "completions/max_terminated_length": 1584.0, "completions/mean_length": 581.87109375, "completions/mean_terminated_length": 581.87109375, "completions/min_length": 147.0, "completions/min_terminated_length": 147.0, "epoch": 0.418177747929259, "frac_reward_zero_std": 1.0, "grad_norm": 0.004137094950964047, "kl": 0.04852294921875, "learning_rate": 5.543463661312847e-07, "loss": 0.0005, "num_tokens": 291955615.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 467 }, { "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": -6.9375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1926.0, "completions/mean_length": 632.26171875, "completions/mean_terminated_length": 621.1141967773438, "completions/min_length": 189.0, "completions/min_terminated_length": 189.0, "epoch": 0.41907320349227667, "frac_reward_zero_std": 0.96875, "grad_norm": 0.06436363037629371, "kl": 0.04949951171875, "learning_rate": 5.512085379471808e-07, "loss": 0.0011, "num_tokens": 292623845.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 468 }, { "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": -6.984375, "completions/max_length": 1338.0, "completions/max_terminated_length": 1338.0, "completions/mean_length": 540.55859375, "completions/mean_terminated_length": 539.2700805664062, "completions/min_length": 200.0, "completions/min_terminated_length": 200.0, "epoch": 0.4199686590552944, "frac_reward_zero_std": 0.96875, "grad_norm": 0.9091175781068817, "kl": 0.0504150390625, "learning_rate": 5.481629666490903e-07, "loss": 0.006, "num_tokens": 293211251.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 469 }, { "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": -7.0, "completions/max_length": 1309.0, "completions/max_terminated_length": 1309.0, "completions/mean_length": 571.876953125, "completions/mean_terminated_length": 571.876953125, "completions/min_length": 209.0, "completions/min_terminated_length": 209.0, "epoch": 0.4208641146183121, "frac_reward_zero_std": 0.96875, "grad_norm": 0.08882810422258497, "kl": 0.04718017578125, "learning_rate": 5.452097800230853e-07, "loss": 0.0039, "num_tokens": 293818340.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 470 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1920.0, "completions/mean_length": 618.740234375, "completions/mean_terminated_length": 615.9432373046875, "completions/min_length": 156.0, "completions/min_terminated_length": 156.0, "epoch": 0.42175957018132976, "frac_reward_zero_std": 0.9375, "grad_norm": 0.08441184034971604, "kl": 0.0458984375, "learning_rate": 5.423491019789623e-07, "loss": 0.0047, "num_tokens": 294451039.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "step": 471 }, { "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": -7.0, "completions/max_length": 1859.0, "completions/max_terminated_length": 1859.0, "completions/mean_length": 571.236328125, "completions/mean_terminated_length": 571.236328125, "completions/min_length": 181.0, "completions/min_terminated_length": 181.0, "epoch": 0.42265502574434743, "frac_reward_zero_std": 1.0, "grad_norm": 0.004151029586595816, "kl": 0.047607421875, "learning_rate": 5.395810525450425e-07, "loss": 0.0005, "num_tokens": 295049912.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 472 }, { "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": -7.0, "completions/max_length": 1622.0, "completions/max_terminated_length": 1622.0, "completions/mean_length": 585.880859375, "completions/mean_terminated_length": 585.880859375, "completions/min_length": 171.0, "completions/min_terminated_length": 171.0, "epoch": 0.4235504813073651, "frac_reward_zero_std": 1.0, "grad_norm": 0.006773189374311078, "kl": 0.0477294921875, "learning_rate": 5.369057478631359e-07, "loss": 0.0005, "num_tokens": 295687771.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 473 }, { "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": -6.96875, "completions/max_length": 1519.0, "completions/max_terminated_length": 1519.0, "completions/mean_length": 559.310546875, "completions/mean_terminated_length": 555.547119140625, "completions/min_length": 141.0, "completions/min_terminated_length": 141.0, "epoch": 0.4244459368703828, "frac_reward_zero_std": 0.96875, "grad_norm": 0.05238104977419995, "kl": 0.0484619140625, "learning_rate": 5.343233001836694e-07, "loss": 0.0077, "num_tokens": 296294970.0, "reward": 0.099609375, "reward_std": 0.0010673906654119492, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "step": 474 }, { "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": -6.953125, "completions/max_length": 2048.0, "completions/max_terminated_length": 1596.0, "completions/mean_length": 597.630859375, "completions/mean_terminated_length": 591.0942993164062, "completions/min_length": 136.0, "completions/min_terminated_length": 136.0, "epoch": 0.42534139243340047, "frac_reward_zero_std": 0.96875, "grad_norm": 0.04626310549442998, "kl": 0.0478515625, "learning_rate": 5.318338178609754e-07, "loss": 0.0124, "num_tokens": 296916957.0, "reward": 0.099609375, "reward_std": 0.0010673906654119492, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "step": 475 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1873.0, "completions/mean_length": 556.599609375, "completions/mean_terminated_length": 550.7510375976562, "completions/min_length": 177.0, "completions/min_terminated_length": 177.0, "epoch": 0.4262368479964182, "frac_reward_zero_std": 1.0, "grad_norm": 0.0053069795960833605, "kl": 0.04669189453125, "learning_rate": 5.294374053487459e-07, "loss": 0.0005, "num_tokens": 297485056.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 476 }, { "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": -7.0, "completions/max_length": 1817.0, "completions/max_terminated_length": 1817.0, "completions/mean_length": 534.4140625, "completions/mean_terminated_length": 534.4140625, "completions/min_length": 167.0, "completions/min_terminated_length": 167.0, "epoch": 0.4271323035594359, "frac_reward_zero_std": 1.0, "grad_norm": 0.004148165474532734, "kl": 0.04693603515625, "learning_rate": 5.271341631956511e-07, "loss": 0.0005, "num_tokens": 298039460.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 477 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1695.0, "completions/mean_length": 564.353515625, "completions/mean_terminated_length": 561.4500732421875, "completions/min_length": 143.0, "completions/min_terminated_length": 143.0, "epoch": 0.42802775912245355, "frac_reward_zero_std": 0.96875, "grad_norm": 0.0361604804039374, "kl": 0.04638671875, "learning_rate": 5.249241880411181e-07, "loss": 0.0153, "num_tokens": 298646553.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 478 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1789.0, "completions/mean_length": 595.029296875, "completions/mean_terminated_length": 592.1859130859375, "completions/min_length": 174.0, "completions/min_terminated_length": 174.0, "epoch": 0.42892321468547123, "frac_reward_zero_std": 0.9375, "grad_norm": 0.07503338366587108, "kl": 0.048583984375, "learning_rate": 5.228075726112785e-07, "loss": 0.0007, "num_tokens": 299286024.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "step": 479 }, { "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": -7.0, "completions/max_length": 1694.0, "completions/max_terminated_length": 1694.0, "completions/mean_length": 589.59375, "completions/mean_terminated_length": 589.59375, "completions/min_length": 173.0, "completions/min_terminated_length": 173.0, "epoch": 0.4298186702484889, "frac_reward_zero_std": 1.0, "grad_norm": 0.003932622151853824, "kl": 0.048095703125, "learning_rate": 5.207844057150768e-07, "loss": 0.0005, "num_tokens": 299916216.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 480 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1600.0, "completions/mean_length": 556.908203125, "completions/mean_terminated_length": 553.990234375, "completions/min_length": 144.0, "completions/min_terminated_length": 144.0, "epoch": 0.4307141258115066, "frac_reward_zero_std": 1.0, "grad_norm": 0.0046160433450423335, "kl": 0.05029296875, "learning_rate": 5.188547722405437e-07, "loss": 0.0005, "num_tokens": 300529225.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 481 }, { "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": -7.0, "completions/max_length": 1835.0, "completions/max_terminated_length": 1835.0, "completions/mean_length": 611.91796875, "completions/mean_terminated_length": 611.91796875, "completions/min_length": 211.0, "completions/min_terminated_length": 211.0, "epoch": 0.43160958137452426, "frac_reward_zero_std": 1.0, "grad_norm": 0.004081618380481847, "kl": 0.04681396484375, "learning_rate": 5.170187531512351e-07, "loss": 0.0005, "num_tokens": 301174607.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 482 }, { "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": -7.0, "completions/max_length": 1538.0, "completions/max_terminated_length": 1538.0, "completions/mean_length": 561.107421875, "completions/mean_terminated_length": 561.107421875, "completions/min_length": 127.0, "completions/min_terminated_length": 127.0, "epoch": 0.432505036937542, "frac_reward_zero_std": 1.0, "grad_norm": 0.005272474133392292, "kl": 0.05059814453125, "learning_rate": 5.152764254828348e-07, "loss": 0.0005, "num_tokens": 301783334.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 483 }, { "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": -7.0, "completions/max_length": 1796.0, "completions/max_terminated_length": 1796.0, "completions/mean_length": 575.103515625, "completions/mean_terminated_length": 575.103515625, "completions/min_length": 176.0, "completions/min_terminated_length": 176.0, "epoch": 0.4334004925005597, "frac_reward_zero_std": 1.0, "grad_norm": 0.004371253746007476, "kl": 0.04931640625, "learning_rate": 5.136278623399225e-07, "loss": 0.0005, "num_tokens": 302405147.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 484 }, { "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": -6.984375, "completions/max_length": 1587.0, "completions/max_terminated_length": 1587.0, "completions/mean_length": 557.376953125, "completions/mean_terminated_length": 555.73779296875, "completions/min_length": 149.0, "completions/min_terminated_length": 149.0, "epoch": 0.43429594806357735, "frac_reward_zero_std": 1.0, "grad_norm": 0.00631760392876256, "kl": 0.04669189453125, "learning_rate": 5.120731328929058e-07, "loss": 0.0005, "num_tokens": 302999388.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 485 }, { "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": -6.984375, "completions/max_length": 1474.0, "completions/max_terminated_length": 1474.0, "completions/mean_length": 579.818359375, "completions/mean_terminated_length": 578.4951171875, "completions/min_length": 201.0, "completions/min_terminated_length": 201.0, "epoch": 0.43519140362659503, "frac_reward_zero_std": 0.96875, "grad_norm": 0.7675240642119238, "kl": 0.05047607421875, "learning_rate": 5.106123023751187e-07, "loss": 0.0069, "num_tokens": 303594319.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 486 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1803.0, "completions/mean_length": 610.142578125, "completions/mean_terminated_length": 604.5039672851562, "completions/min_length": 199.0, "completions/min_terminated_length": 199.0, "epoch": 0.4360868591896127, "frac_reward_zero_std": 1.0, "grad_norm": 0.003966772271342612, "kl": 0.04791259765625, "learning_rate": 5.092454320800833e-07, "loss": 0.0005, "num_tokens": 304248552.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 487 }, { "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": -6.953125, "completions/max_length": 2048.0, "completions/max_terminated_length": 1912.0, "completions/mean_length": 583.4921875, "completions/mean_terminated_length": 574.8605346679688, "completions/min_length": 159.0, "completions/min_terminated_length": 159.0, "epoch": 0.4369823147526304, "frac_reward_zero_std": 0.96875, "grad_norm": 0.06718584348509721, "kl": 0.04815673828125, "learning_rate": 5.079725793589405e-07, "loss": 0.0119, "num_tokens": 304857716.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 488 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1631.0, "completions/mean_length": 587.619140625, "completions/mean_terminated_length": 584.76123046875, "completions/min_length": 151.0, "completions/min_terminated_length": 151.0, "epoch": 0.43787777031564806, "frac_reward_zero_std": 0.96875, "grad_norm": 0.06044390352276465, "kl": 0.04400634765625, "learning_rate": 5.067937976180407e-07, "loss": 0.0005, "num_tokens": 305463137.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 489 }, { "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": -7.0, "completions/max_length": 1466.0, "completions/max_terminated_length": 1466.0, "completions/mean_length": 587.826171875, "completions/mean_terminated_length": 587.826171875, "completions/min_length": 148.0, "completions/min_terminated_length": 148.0, "epoch": 0.4387732258786658, "frac_reward_zero_std": 1.0, "grad_norm": 0.004340416142239811, "kl": 0.04534912109375, "learning_rate": 5.057091363167046e-07, "loss": 0.0005, "num_tokens": 306081032.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 490 }, { "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": -7.0, "completions/max_length": 1335.0, "completions/max_terminated_length": 1335.0, "completions/mean_length": 544.626953125, "completions/mean_terminated_length": 544.626953125, "completions/min_length": 165.0, "completions/min_terminated_length": 165.0, "epoch": 0.4396686814416835, "frac_reward_zero_std": 1.0, "grad_norm": 0.005717450638553965, "kl": 0.0474853515625, "learning_rate": 5.047186409651489e-07, "loss": 0.0005, "num_tokens": 306656489.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 491 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1721.0, "completions/mean_length": 613.29296875, "completions/mean_terminated_length": 607.6666870117188, "completions/min_length": 170.0, "completions/min_terminated_length": 170.0, "epoch": 0.44056413700470115, "frac_reward_zero_std": 0.96875, "grad_norm": 0.05273242178002334, "kl": 0.0477294921875, "learning_rate": 5.038223531225742e-07, "loss": 0.0115, "num_tokens": 307264911.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 492 }, { "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": -7.0, "completions/max_length": 1694.0, "completions/max_terminated_length": 1694.0, "completions/mean_length": 554.16796875, "completions/mean_terminated_length": 554.16796875, "completions/min_length": 117.0, "completions/min_terminated_length": 117.0, "epoch": 0.44145959256771883, "frac_reward_zero_std": 1.0, "grad_norm": 0.004061493509151464, "kl": 0.04656982421875, "learning_rate": 5.030203103954232e-07, "loss": 0.0005, "num_tokens": 307848181.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 493 }, { "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": -7.0, "completions/max_length": 1524.0, "completions/max_terminated_length": 1524.0, "completions/mean_length": 558.9453125, "completions/mean_terminated_length": 558.9453125, "completions/min_length": 161.0, "completions/min_terminated_length": 161.0, "epoch": 0.4423550481307365, "frac_reward_zero_std": 1.0, "grad_norm": 0.004367225471207651, "kl": 0.04486083984375, "learning_rate": 5.023125464358026e-07, "loss": 0.0004, "num_tokens": 308417529.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 494 }, { "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": -7.0, "completions/max_length": 1677.0, "completions/max_terminated_length": 1677.0, "completions/mean_length": 551.27734375, "completions/mean_terminated_length": 551.27734375, "completions/min_length": 133.0, "completions/min_terminated_length": 133.0, "epoch": 0.4432505036937542, "frac_reward_zero_std": 1.0, "grad_norm": 0.00418532184825478, "kl": 0.047119140625, "learning_rate": 5.016990909400709e-07, "loss": 0.0005, "num_tokens": 308975255.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 495 }, { "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": -7.0, "completions/max_length": 1497.0, "completions/max_terminated_length": 1497.0, "completions/mean_length": 601.00390625, "completions/mean_terminated_length": 601.00390625, "completions/min_length": 179.0, "completions/min_terminated_length": 179.0, "epoch": 0.44414595925677186, "frac_reward_zero_std": 1.0, "grad_norm": 0.004602536366235129, "kl": 0.04852294921875, "learning_rate": 5.011799696475915e-07, "loss": 0.0005, "num_tokens": 309577561.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 496 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1624.0, "completions/mean_length": 531.177734375, "completions/mean_terminated_length": 528.2094116210938, "completions/min_length": 181.0, "completions/min_terminated_length": 181.0, "epoch": 0.4450414148197896, "frac_reward_zero_std": 1.0, "grad_norm": 0.004030212878424051, "kl": 0.04644775390625, "learning_rate": 5.007552043396547e-07, "loss": 0.0005, "num_tokens": 310151220.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 497 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1416.0, "completions/mean_length": 545.96875, "completions/mean_terminated_length": 543.0293579101562, "completions/min_length": 152.0, "completions/min_terminated_length": 152.0, "epoch": 0.44593687038280727, "frac_reward_zero_std": 0.96875, "grad_norm": 0.04970801239600626, "kl": 0.047119140625, "learning_rate": 5.004248128385618e-07, "loss": -0.0006, "num_tokens": 310755700.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 498 }, { "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": -7.0, "completions/max_length": 1445.0, "completions/max_terminated_length": 1445.0, "completions/mean_length": 556.638671875, "completions/mean_terminated_length": 556.638671875, "completions/min_length": 185.0, "completions/min_terminated_length": 185.0, "epoch": 0.44683232594582495, "frac_reward_zero_std": 1.0, "grad_norm": 0.004571033839926497, "kl": 0.0477294921875, "learning_rate": 5.001888090068784e-07, "loss": 0.0005, "num_tokens": 311324955.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 499 }, { "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": -7.0, "completions/max_length": 1470.0, "completions/max_terminated_length": 1470.0, "completions/mean_length": 580.17578125, "completions/mean_terminated_length": 580.17578125, "completions/min_length": 204.0, "completions/min_terminated_length": 204.0, "epoch": 0.4477277815088426, "frac_reward_zero_std": 1.0, "grad_norm": 0.004267414031586906, "kl": 0.04730224609375, "learning_rate": 5.000472027468528e-07, "loss": 0.0005, "num_tokens": 311967237.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 500 }, { "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": -7.0, "completions/max_length": 1609.0, "completions/max_terminated_length": 1609.0, "completions/mean_length": 606.591796875, "completions/mean_terminated_length": 606.591796875, "completions/min_length": 146.0, "completions/min_terminated_length": 146.0, "epoch": 0.4486232370718603, "frac_reward_zero_std": 0.96875, "grad_norm": 0.05086077162905058, "kl": 0.0474853515625, "learning_rate": 5.000000000000001e-07, "loss": 0.0011, "num_tokens": 312581300.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 501 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1779.0, "completions/mean_length": 599.373046875, "completions/mean_terminated_length": 596.5381469726562, "completions/min_length": 108.0, "completions/min_terminated_length": 108.0, "epoch": 0.449518692634878, "frac_reward_zero_std": 0.96875, "grad_norm": 0.056542864188258934, "kl": 0.0482177734375, "learning_rate": 5.000472027468528e-07, "loss": 0.0146, "num_tokens": 313236387.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 502 }, { "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": -7.0, "completions/max_length": 1390.0, "completions/max_terminated_length": 1390.0, "completions/mean_length": 565.26171875, "completions/mean_terminated_length": 565.26171875, "completions/min_length": 186.0, "completions/min_terminated_length": 186.0, "epoch": 0.45041414819789566, "frac_reward_zero_std": 1.0, "grad_norm": 0.003944884798198514, "kl": 0.04736328125, "learning_rate": 5.001888090068784e-07, "loss": 0.0005, "num_tokens": 313811833.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 503 }, { "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": -7.0, "completions/max_length": 1529.0, "completions/max_terminated_length": 1529.0, "completions/mean_length": 568.341796875, "completions/mean_terminated_length": 568.341796875, "completions/min_length": 203.0, "completions/min_terminated_length": 203.0, "epoch": 0.4513096037609134, "frac_reward_zero_std": 1.0, "grad_norm": 0.005105667807983009, "kl": 0.0458984375, "learning_rate": 5.004248128385618e-07, "loss": 0.0005, "num_tokens": 314399896.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 504 }, { "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": -7.0, "completions/max_length": 1788.0, "completions/max_terminated_length": 1788.0, "completions/mean_length": 609.244140625, "completions/mean_terminated_length": 609.244140625, "completions/min_length": 175.0, "completions/min_terminated_length": 175.0, "epoch": 0.45220505932393107, "frac_reward_zero_std": 1.0, "grad_norm": 0.0043507543970794085, "kl": 0.0474853515625, "learning_rate": 5.007552043396547e-07, "loss": 0.0005, "num_tokens": 315061797.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 505 }, { "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": -6.984375, "completions/max_length": 2048.0, "completions/max_terminated_length": 1824.0, "completions/mean_length": 574.73828125, "completions/mean_terminated_length": 571.8551635742188, "completions/min_length": 139.0, "completions/min_terminated_length": 139.0, "epoch": 0.45310051488694875, "frac_reward_zero_std": 0.96875, "grad_norm": 0.08176921037298442, "kl": 0.0460205078125, "learning_rate": 5.011799696475915e-07, "loss": 0.0138, "num_tokens": 315676863.0, "reward": 0.09980468451976776, "reward_std": 0.0007812500116415322, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.998046875, "rewards/format_reward/std": 0.04419417306780815, "step": 506 }, { "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": -6.96875, "completions/max_length": 2048.0, "completions/max_terminated_length": 1696.0, "completions/mean_length": 594.41796875, "completions/mean_terminated_length": 588.7176513671875, "completions/min_length": 176.0, "completions/min_terminated_length": 176.0, "epoch": 0.4539959704499664, "frac_reward_zero_std": 1.0, "grad_norm": 0.004202191868376494, "kl": 0.04669189453125, "learning_rate": 5.016990909400706e-07, "loss": 0.0005, "num_tokens": 316299381.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 507 }, { "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": -7.0, "completions/max_length": 1558.0, "completions/max_terminated_length": 1558.0, "completions/mean_length": 567.14453125, "completions/mean_terminated_length": 567.14453125, "completions/min_length": 209.0, "completions/min_terminated_length": 209.0, "epoch": 0.4548914260129841, "frac_reward_zero_std": 0.9375, "grad_norm": 0.06400276605298859, "kl": 0.0452880859375, "learning_rate": 5.023125464358026e-07, "loss": -0.0009, "num_tokens": 316861775.0, "reward": 0.099609375, "reward_std": 0.0015625000232830644, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 0.99609375, "rewards/format_reward/std": 0.06243881583213806, "step": 508 }, { "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": -7.0, "completions/max_length": 1631.0, "completions/max_terminated_length": 1631.0, "completions/mean_length": 544.9765625, "completions/mean_terminated_length": 544.9765625, "completions/min_length": 194.0, "completions/min_terminated_length": 194.0, "epoch": 0.4557868815760018, "frac_reward_zero_std": 1.0, "grad_norm": 0.004374618149752765, "kl": 0.04608154296875, "learning_rate": 5.03020310395423e-07, "loss": 0.0005, "num_tokens": 317450899.0, "reward": 0.10000000149011612, "reward_std": 0.0, "rewards/code_reward/mean": 0.0, "rewards/code_reward/std": 0.0, "rewards/format_reward/mean": 1.0, "rewards/format_reward/std": 0.0, "step": 509 }, { "epoch": 0.4557868815760018, "step": 509, "total_flos": 0.0, "train_loss": 9.052564064315237e-07, "train_runtime": 75.296, "train_samples_per_second": 3399.916, "train_steps_per_second": 6.64 } ], "logging_steps": 1, "max_steps": 500, "num_input_tokens_seen": 317450899, "num_train_epochs": 1, "save_steps": 50, "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": 16, "trial_name": null, "trial_params": null }