| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9921259842519685, | |
| "eval_steps": 500, | |
| "global_step": 42, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 14389.429036458334, | |
| "epoch": 0.023622047244094488, | |
| "grad_norm": 0.18101558089256287, | |
| "kl": 0.0, | |
| "learning_rate": 6.000000000000001e-07, | |
| "loss": 0.0, | |
| "reward": 0.09523809949556987, | |
| "reward_std": 0.2519763112068176, | |
| "rewards/accuracy_reward": 0.09523809949556987, | |
| "step": 1 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 14440.095703125, | |
| "epoch": 0.047244094488188976, | |
| "grad_norm": 0.10683715343475342, | |
| "kl": 0.0, | |
| "learning_rate": 1.2000000000000002e-06, | |
| "loss": -0.0, | |
| "reward": 0.09523809949556987, | |
| "reward_std": 0.16265000899632773, | |
| "rewards/accuracy_reward": 0.09523809949556987, | |
| "step": 2 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 14900.953125, | |
| "epoch": 0.07086614173228346, | |
| "grad_norm": 0.09402324259281158, | |
| "kl": 9.489059448242188e-05, | |
| "learning_rate": 1.8e-06, | |
| "loss": 0.0, | |
| "reward": 0.047619049747784935, | |
| "reward_std": 0.1259881556034088, | |
| "rewards/accuracy_reward": 0.047619049747784935, | |
| "step": 3 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 12512.096028645834, | |
| "epoch": 0.09448818897637795, | |
| "grad_norm": 0.20009347796440125, | |
| "kl": 0.0001041094462076823, | |
| "learning_rate": 2.4000000000000003e-06, | |
| "loss": 0.0, | |
| "reward": 0.1428571492433548, | |
| "reward_std": 0.2886381645997365, | |
| "rewards/accuracy_reward": 0.1428571492433548, | |
| "step": 4 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 15601.238932291666, | |
| "epoch": 0.11811023622047244, | |
| "grad_norm": 0.0005439579836092889, | |
| "kl": 7.62939453125e-05, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0, | |
| "reward": 0.0, | |
| "reward_std": 0.0, | |
| "rewards/accuracy_reward": 0.0, | |
| "step": 5 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 14982.953125, | |
| "epoch": 0.14173228346456693, | |
| "grad_norm": 0.17226731777191162, | |
| "kl": 8.336702982584636e-05, | |
| "learning_rate": 2.99459623281379e-06, | |
| "loss": 0.0, | |
| "reward": 0.09523809949556987, | |
| "reward_std": 0.2519763112068176, | |
| "rewards/accuracy_reward": 0.09523809949556987, | |
| "step": 6 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 14418.096028645834, | |
| "epoch": 0.16535433070866143, | |
| "grad_norm": 0.2076222449541092, | |
| "kl": 9.473164876302083e-05, | |
| "learning_rate": 2.978423865521563e-06, | |
| "loss": 0.0, | |
| "reward": 0.2857143034537633, | |
| "reward_std": 0.4668123225371043, | |
| "rewards/accuracy_reward": 0.2857143034537633, | |
| "step": 7 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 13729.191080729166, | |
| "epoch": 0.1889763779527559, | |
| "grad_norm": 0.13402122259140015, | |
| "kl": 0.00010665257771809895, | |
| "learning_rate": 2.9515994204002487e-06, | |
| "loss": 0.0, | |
| "reward": 0.14285715421040854, | |
| "reward_std": 0.17817415793736777, | |
| "rewards/accuracy_reward": 0.14285715421040854, | |
| "step": 8 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 12956.857747395834, | |
| "epoch": 0.2125984251968504, | |
| "grad_norm": 0.11447420716285706, | |
| "kl": 0.00014670689900716147, | |
| "learning_rate": 2.9143161681916264e-06, | |
| "loss": 0.0, | |
| "reward": 0.047619049747784935, | |
| "reward_std": 0.1259881556034088, | |
| "rewards/accuracy_reward": 0.047619049747784935, | |
| "step": 9 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 11794.762532552084, | |
| "epoch": 0.23622047244094488, | |
| "grad_norm": 0.0022508089896291494, | |
| "kl": 0.00014130274454752603, | |
| "learning_rate": 2.866842735582204e-06, | |
| "loss": 0.0, | |
| "reward": 0.0, | |
| "reward_std": 0.0, | |
| "rewards/accuracy_reward": 0.0, | |
| "step": 10 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 12350.71484375, | |
| "epoch": 0.25984251968503935, | |
| "grad_norm": 0.19783379137516022, | |
| "kl": 0.00015211105346679688, | |
| "learning_rate": 2.8095211697417823e-06, | |
| "loss": 0.0, | |
| "reward": 0.09523809949556987, | |
| "reward_std": 0.2519763112068176, | |
| "rewards/accuracy_reward": 0.09523809949556987, | |
| "step": 11 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 15620.238606770834, | |
| "epoch": 0.28346456692913385, | |
| "grad_norm": 0.004334005527198315, | |
| "kl": 0.0004711151123046875, | |
| "learning_rate": 2.7427644738657634e-06, | |
| "loss": 0.0, | |
| "reward": 0.0, | |
| "reward_std": 0.0, | |
| "rewards/accuracy_reward": 0.0, | |
| "step": 12 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 12504.857747395834, | |
| "epoch": 0.30708661417322836, | |
| "grad_norm": 0.23003408312797546, | |
| "kl": 0.00032520294189453125, | |
| "learning_rate": 2.6670536314776595e-06, | |
| "loss": 0.0, | |
| "reward": 0.2380952537059784, | |
| "reward_std": 0.43015046914418537, | |
| "rewards/accuracy_reward": 0.2380952537059784, | |
| "step": 13 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 15596.096028645834, | |
| "epoch": 0.33070866141732286, | |
| "grad_norm": 0.0899478867650032, | |
| "kl": 0.0004437764485677083, | |
| "learning_rate": 2.5829341409317867e-06, | |
| "loss": 0.0, | |
| "reward": 0.047619049747784935, | |
| "reward_std": 0.1259881556034088, | |
| "rewards/accuracy_reward": 0.047619049747784935, | |
| "step": 14 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 12472.667317708334, | |
| "epoch": 0.3543307086614173, | |
| "grad_norm": 0.1614641547203064, | |
| "kl": 0.00022761027018229166, | |
| "learning_rate": 2.4910120850851222e-06, | |
| "loss": 0.0, | |
| "reward": 0.047619049747784935, | |
| "reward_std": 0.1259881556034088, | |
| "rewards/accuracy_reward": 0.047619049747784935, | |
| "step": 15 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 14462.953125, | |
| "epoch": 0.3779527559055118, | |
| "grad_norm": 0.1555420160293579, | |
| "kl": 0.0011622111002604167, | |
| "learning_rate": 2.39194976445643e-06, | |
| "loss": 0.0, | |
| "reward": 0.19047619899113974, | |
| "reward_std": 0.32530001799265545, | |
| "rewards/accuracy_reward": 0.19047619899113974, | |
| "step": 16 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 14019.333984375, | |
| "epoch": 0.4015748031496063, | |
| "grad_norm": 0.006728035863488913, | |
| "kl": 0.0006306966145833334, | |
| "learning_rate": 2.286460925335848e-06, | |
| "loss": 0.0, | |
| "reward": 0.0, | |
| "reward_std": 0.0, | |
| "rewards/accuracy_reward": 0.0, | |
| "step": 17 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 14344.667317708334, | |
| "epoch": 0.4251968503937008, | |
| "grad_norm": 0.10385935008525848, | |
| "kl": 0.0008605321248372396, | |
| "learning_rate": 2.17530561722651e-06, | |
| "loss": 0.0, | |
| "reward": 0.14285715421040854, | |
| "reward_std": 0.17817415793736777, | |
| "rewards/accuracy_reward": 0.14285715421040854, | |
| "step": 18 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 15407.238932291666, | |
| "epoch": 0.44881889763779526, | |
| "grad_norm": 0.007518239319324493, | |
| "kl": 0.0008875528971354166, | |
| "learning_rate": 2.059284716670463e-06, | |
| "loss": 0.0, | |
| "reward": 0.0, | |
| "reward_std": 0.0, | |
| "rewards/accuracy_reward": 0.0, | |
| "step": 19 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 11890.9052734375, | |
| "epoch": 0.47244094488188976, | |
| "grad_norm": 0.15603958070278168, | |
| "kl": 0.0013834635416666667, | |
| "learning_rate": 1.9392341569148255e-06, | |
| "loss": 0.0001, | |
| "reward": 0.23809524377187094, | |
| "reward_std": 0.16265000899632773, | |
| "rewards/accuracy_reward": 0.23809524377187094, | |
| "step": 20 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 13337.333658854166, | |
| "epoch": 0.49606299212598426, | |
| "grad_norm": 0.007209372241050005, | |
| "kl": 0.0018170674641927083, | |
| "learning_rate": 1.8160189049935894e-06, | |
| "loss": 0.0001, | |
| "reward": 0.3333333333333333, | |
| "reward_std": 0.0, | |
| "rewards/accuracy_reward": 0.3333333333333333, | |
| "step": 21 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 14147.5244140625, | |
| "epoch": 0.5196850393700787, | |
| "grad_norm": 0.16794154047966003, | |
| "kl": 0.0015004475911458333, | |
| "learning_rate": 1.6905267296203183e-06, | |
| "loss": 0.0001, | |
| "reward": 0.3333333432674408, | |
| "reward_std": 0.32530001799265545, | |
| "rewards/accuracy_reward": 0.3333333432674408, | |
| "step": 22 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 10586.619303385416, | |
| "epoch": 0.5433070866141733, | |
| "grad_norm": 0.24202802777290344, | |
| "kl": 0.0016371409098307292, | |
| "learning_rate": 1.5636618047942224e-06, | |
| "loss": 0.0001, | |
| "reward": 0.571428601940473, | |
| "reward_std": 0.4146263202031453, | |
| "rewards/accuracy_reward": 0.571428601940473, | |
| "step": 23 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 12914.190755208334, | |
| "epoch": 0.5669291338582677, | |
| "grad_norm": 0.006770447362214327, | |
| "kl": 0.0011011759440104167, | |
| "learning_rate": 1.4363381952057778e-06, | |
| "loss": 0.0, | |
| "reward": 0.0, | |
| "reward_std": 0.0, | |
| "rewards/accuracy_reward": 0.0, | |
| "step": 24 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 15026.238606770834, | |
| "epoch": 0.5905511811023622, | |
| "grad_norm": 0.14896388351917267, | |
| "kl": 0.0005785624186197916, | |
| "learning_rate": 1.3094732703796818e-06, | |
| "loss": 0.0, | |
| "reward": 0.19047620395819345, | |
| "reward_std": 0.30416231354077655, | |
| "rewards/accuracy_reward": 0.19047620395819345, | |
| "step": 25 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 14715.4765625, | |
| "epoch": 0.6141732283464567, | |
| "grad_norm": 0.00993309449404478, | |
| "kl": 0.001087188720703125, | |
| "learning_rate": 1.183981095006411e-06, | |
| "loss": 0.0, | |
| "reward": 0.0, | |
| "reward_std": 0.0, | |
| "rewards/accuracy_reward": 0.0, | |
| "step": 26 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 12330.762532552084, | |
| "epoch": 0.6377952755905512, | |
| "grad_norm": 0.008712113834917545, | |
| "kl": 0.0025761922200520835, | |
| "learning_rate": 1.0607658430851746e-06, | |
| "loss": 0.0001, | |
| "reward": 0.0, | |
| "reward_std": 0.0, | |
| "rewards/accuracy_reward": 0.0, | |
| "step": 27 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 13653.619140625, | |
| "epoch": 0.6614173228346457, | |
| "grad_norm": 0.007787778973579407, | |
| "kl": 0.0022443135579427085, | |
| "learning_rate": 9.407152833295372e-07, | |
| "loss": 0.0001, | |
| "reward": 0.0, | |
| "reward_std": 0.0, | |
| "rewards/accuracy_reward": 0.0, | |
| "step": 28 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 13795.381184895834, | |
| "epoch": 0.6850393700787402, | |
| "grad_norm": 0.008136502467095852, | |
| "kl": 0.0014813741048177083, | |
| "learning_rate": 8.246943827734898e-07, | |
| "loss": 0.0001, | |
| "reward": 0.0, | |
| "reward_std": 0.0, | |
| "rewards/accuracy_reward": 0.0, | |
| "step": 29 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 15744.857747395834, | |
| "epoch": 0.7086614173228346, | |
| "grad_norm": 0.005922618322074413, | |
| "kl": 0.0023244222005208335, | |
| "learning_rate": 7.135390746641527e-07, | |
| "loss": 0.0001, | |
| "reward": 0.0, | |
| "reward_std": 0.0, | |
| "rewards/accuracy_reward": 0.0, | |
| "step": 30 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 12687.0009765625, | |
| "epoch": 0.7322834645669292, | |
| "grad_norm": 0.10701826214790344, | |
| "kl": 0.0008004506429036459, | |
| "learning_rate": 6.080502355435701e-07, | |
| "loss": 0.0, | |
| "reward": 0.09523809949556987, | |
| "reward_std": 0.16265000899632773, | |
| "rewards/accuracy_reward": 0.09523809949556987, | |
| "step": 31 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 13935.095703125, | |
| "epoch": 0.7559055118110236, | |
| "grad_norm": 0.10233797878026962, | |
| "kl": 0.002204259236653646, | |
| "learning_rate": 5.089879149148781e-07, | |
| "loss": 0.0001, | |
| "reward": 0.14285715421040854, | |
| "reward_std": 0.17817415793736777, | |
| "rewards/accuracy_reward": 0.14285715421040854, | |
| "step": 32 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 15292.429361979166, | |
| "epoch": 0.7795275590551181, | |
| "grad_norm": 0.1071338877081871, | |
| "kl": 0.0007845560709635416, | |
| "learning_rate": 4.170658590682134e-07, | |
| "loss": 0.0, | |
| "reward": 0.14285715421040854, | |
| "reward_std": 0.17817415793736777, | |
| "rewards/accuracy_reward": 0.14285715421040854, | |
| "step": 33 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 12384.1435546875, | |
| "epoch": 0.8031496062992126, | |
| "grad_norm": 0.19518683850765228, | |
| "kl": 0.0014890034993489583, | |
| "learning_rate": 3.3294636852234106e-07, | |
| "loss": 0.0001, | |
| "reward": 0.3809523979822795, | |
| "reward_std": 0.35634831587473553, | |
| "rewards/accuracy_reward": 0.3809523979822795, | |
| "step": 34 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 15084.381510416666, | |
| "epoch": 0.8267716535433071, | |
| "grad_norm": 0.12067427486181259, | |
| "kl": 0.0007756551106770834, | |
| "learning_rate": 2.572355261342369e-07, | |
| "loss": 0.0, | |
| "reward": 0.09523809949556987, | |
| "reward_std": 0.16265000899632773, | |
| "rewards/accuracy_reward": 0.09523809949556987, | |
| "step": 35 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 13533.048502604166, | |
| "epoch": 0.8503937007874016, | |
| "grad_norm": 0.1780463457107544, | |
| "kl": 0.0023701985677083335, | |
| "learning_rate": 1.9047883025821777e-07, | |
| "loss": 0.0001, | |
| "reward": 0.3333333432674408, | |
| "reward_std": 0.32530001799265545, | |
| "rewards/accuracy_reward": 0.3333333432674408, | |
| "step": 36 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 12354.524088541666, | |
| "epoch": 0.8740157480314961, | |
| "grad_norm": 0.008607493713498116, | |
| "kl": 0.0020573933919270835, | |
| "learning_rate": 1.3315726441779629e-07, | |
| "loss": 0.0001, | |
| "reward": 0.0, | |
| "reward_std": 0.0, | |
| "rewards/accuracy_reward": 0.0, | |
| "step": 37 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 14624.238606770834, | |
| "epoch": 0.8976377952755905, | |
| "grad_norm": 0.00765492208302021, | |
| "kl": 0.0015093485514322917, | |
| "learning_rate": 8.568383180837369e-08, | |
| "loss": 0.0001, | |
| "reward": 0.0, | |
| "reward_std": 0.0, | |
| "rewards/accuracy_reward": 0.0, | |
| "step": 38 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 14310.810221354166, | |
| "epoch": 0.9212598425196851, | |
| "grad_norm": 0.009644408710300922, | |
| "kl": 0.0020573933919270835, | |
| "learning_rate": 4.8400579599751696e-08, | |
| "loss": 0.0001, | |
| "reward": 0.0, | |
| "reward_std": 0.0, | |
| "rewards/accuracy_reward": 0.0, | |
| "step": 39 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 11984.619547526041, | |
| "epoch": 0.9448818897637795, | |
| "grad_norm": 0.008329860866069794, | |
| "kl": 0.0015360514322916667, | |
| "learning_rate": 2.1576134478437316e-08, | |
| "loss": 0.0001, | |
| "reward": 0.0, | |
| "reward_std": 0.0, | |
| "rewards/accuracy_reward": 0.0, | |
| "step": 40 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 13679.857747395834, | |
| "epoch": 0.968503937007874, | |
| "grad_norm": 0.16029423475265503, | |
| "kl": 0.001178741455078125, | |
| "learning_rate": 5.403767186210218e-09, | |
| "loss": 0.0, | |
| "reward": 0.047619049747784935, | |
| "reward_std": 0.1259881556034088, | |
| "rewards/accuracy_reward": 0.047619049747784935, | |
| "step": 41 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 12962.333333333334, | |
| "epoch": 0.9921259842519685, | |
| "grad_norm": 0.010274315252900124, | |
| "kl": 0.0017293294270833333, | |
| "learning_rate": 0.0, | |
| "loss": 0.0001, | |
| "reward": 0.0, | |
| "reward_std": 0.0, | |
| "rewards/accuracy_reward": 0.0, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.9921259842519685, | |
| "step": 42, | |
| "total_flos": 0.0, | |
| "train_loss": 4.045824278607414e-05, | |
| "train_runtime": 33605.9584, | |
| "train_samples_per_second": 0.004, | |
| "train_steps_per_second": 0.001 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 42, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |