{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 190,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 511.59375,
      "epoch": 0.02631578947368421,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 9.973684210526315e-07,
      "loss": 0.0,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1
    },
    {
      "completion_length": 506.4375,
      "epoch": 0.05263157894736842,
      "grad_norm": 1.1859753641586046,
      "kl": 0.0,
      "learning_rate": 9.947368421052631e-07,
      "loss": 0.0,
      "reward": 1.90625,
      "reward_std": 0.13466878235340118,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 2
    },
    {
      "completion_length": 510.25,
      "epoch": 0.07894736842105263,
      "grad_norm": 0.01741859728971303,
      "kl": 0.00041961669921875,
      "learning_rate": 9.921052631578947e-07,
      "loss": 0.0,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 0.96875,
      "step": 3
    },
    {
      "completion_length": 500.78125,
      "epoch": 0.10526315789473684,
      "grad_norm": 1.5842460584085565,
      "kl": 0.0003948211669921875,
      "learning_rate": 9.894736842105263e-07,
      "loss": 0.0,
      "reward": 1.75,
      "reward_std": 0.19716878235340118,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 4
    },
    {
      "completion_length": 513.125,
      "epoch": 0.13157894736842105,
      "grad_norm": 0.018266102639759176,
      "kl": 0.00038909912109375,
      "learning_rate": 9.868421052631579e-07,
      "loss": 0.0,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 5
    },
    {
      "completion_length": 513.5,
      "epoch": 0.15789473684210525,
      "grad_norm": 0.013048600742559903,
      "kl": 0.0004444122314453125,
      "learning_rate": 9.842105263157894e-07,
      "loss": 0.0,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 6
    },
    {
      "completion_length": 510.625,
      "epoch": 0.18421052631578946,
      "grad_norm": 1.2716482982161232,
      "kl": 0.000408172607421875,
      "learning_rate": 9.81578947368421e-07,
      "loss": 0.0,
      "reward": 1.90625,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 7
    },
    {
      "completion_length": 513.0,
      "epoch": 0.21052631578947367,
      "grad_norm": 0.01666083229438018,
      "kl": 0.0004787445068359375,
      "learning_rate": 9.789473684210526e-07,
      "loss": 0.0,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 8
    },
    {
      "completion_length": 503.65625,
      "epoch": 0.23684210526315788,
      "grad_norm": 1.5384360001027126,
      "kl": 0.0005340576171875,
      "learning_rate": 9.763157894736842e-07,
      "loss": 0.0,
      "reward": 1.8125,
      "reward_std": 0.21650634706020355,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 9
    },
    {
      "completion_length": 497.0,
      "epoch": 0.2631578947368421,
      "grad_norm": 0.8586709618188666,
      "kl": 0.0004787445068359375,
      "learning_rate": 9.736842105263158e-07,
      "loss": 0.0,
      "reward": 1.90625,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 10
    },
    {
      "completion_length": 500.0,
      "epoch": 0.2894736842105263,
      "grad_norm": 3.3263216841342786,
      "kl": 0.000499725341796875,
      "learning_rate": 9.710526315789474e-07,
      "loss": 0.0,
      "reward": 1.9375,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 11
    },
    {
      "completion_length": 478.8125,
      "epoch": 0.3157894736842105,
      "grad_norm": 1.4312975247602437,
      "kl": 0.000457763671875,
      "learning_rate": 9.68421052631579e-07,
      "loss": 0.0,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 12
    },
    {
      "completion_length": 521.71875,
      "epoch": 0.34210526315789475,
      "grad_norm": 0.9553780876367552,
      "kl": 0.0006103515625,
      "learning_rate": 9.657894736842105e-07,
      "loss": 0.0,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 13
    },
    {
      "completion_length": 479.21875,
      "epoch": 0.3684210526315789,
      "grad_norm": 0.04637496251159502,
      "kl": 0.000560760498046875,
      "learning_rate": 9.63157894736842e-07,
      "loss": 0.0,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 14
    },
    {
      "completion_length": 497.9375,
      "epoch": 0.39473684210526316,
      "grad_norm": 0.016384627376893796,
      "kl": 0.000553131103515625,
      "learning_rate": 9.605263157894737e-07,
      "loss": 0.0,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 15
    },
    {
      "completion_length": 503.09375,
      "epoch": 0.42105263157894735,
      "grad_norm": 4.266561310972031,
      "kl": 0.000637054443359375,
      "learning_rate": 9.578947368421053e-07,
      "loss": 0.0,
      "reward": 1.875,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 16
    },
    {
      "completion_length": 524.1875,
      "epoch": 0.4473684210526316,
      "grad_norm": 1.5312023004868032,
      "kl": 0.00067901611328125,
      "learning_rate": 9.552631578947368e-07,
      "loss": 0.0,
      "reward": 1.9375,
      "reward_std": 0.07216878235340118,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 17
    },
    {
      "completion_length": 502.40625,
      "epoch": 0.47368421052631576,
      "grad_norm": 0.03266540421884785,
      "kl": 0.00070953369140625,
      "learning_rate": 9.526315789473683e-07,
      "loss": 0.0,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 18
    },
    {
      "completion_length": 492.34375,
      "epoch": 0.5,
      "grad_norm": 1.657170732210159,
      "kl": 0.000843048095703125,
      "learning_rate": 9.499999999999999e-07,
      "loss": 0.0,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.96875,
      "step": 19
    },
    {
      "completion_length": 490.3125,
      "epoch": 0.5263157894736842,
      "grad_norm": 0.026300116989117366,
      "kl": 0.00086212158203125,
      "learning_rate": 9.473684210526315e-07,
      "loss": 0.0,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 20
    },
    {
      "completion_length": 496.875,
      "epoch": 0.5526315789473685,
      "grad_norm": 1.0089214071204007,
      "kl": 0.000843048095703125,
      "learning_rate": 9.447368421052632e-07,
      "loss": 0.0,
      "reward": 1.9375,
      "reward_std": 0.07216878235340118,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 21
    },
    {
      "completion_length": 505.3125,
      "epoch": 0.5789473684210527,
      "grad_norm": 2.2869920085637188,
      "kl": 0.00106048583984375,
      "learning_rate": 9.421052631578948e-07,
      "loss": 0.0,
      "reward": 1.84375,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 22
    },
    {
      "completion_length": 541.65625,
      "epoch": 0.6052631578947368,
      "grad_norm": 0.6352260789447539,
      "kl": 0.00102996826171875,
      "learning_rate": 9.394736842105263e-07,
      "loss": 0.0,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.96875,
      "step": 23
    },
    {
      "completion_length": 514.125,
      "epoch": 0.631578947368421,
      "grad_norm": 0.020950649700612674,
      "kl": 0.00090789794921875,
      "learning_rate": 9.368421052631579e-07,
      "loss": 0.0,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 24
    },
    {
      "completion_length": 508.875,
      "epoch": 0.6578947368421053,
      "grad_norm": 0.02679766826877623,
      "kl": 0.001007080078125,
      "learning_rate": 9.342105263157895e-07,
      "loss": 0.0,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 25
    },
    {
      "completion_length": 512.8125,
      "epoch": 0.6842105263157895,
      "grad_norm": 0.9944506905480381,
      "kl": 0.00101470947265625,
      "learning_rate": 9.31578947368421e-07,
      "loss": 0.0,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 26
    },
    {
      "completion_length": 506.875,
      "epoch": 0.7105263157894737,
      "grad_norm": 0.968143218853605,
      "kl": 0.00107574462890625,
      "learning_rate": 9.289473684210526e-07,
      "loss": 0.0,
      "reward": 1.90625,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 27
    },
    {
      "completion_length": 485.5,
      "epoch": 0.7368421052631579,
      "grad_norm": 1.0987810816945758,
      "kl": 0.001129150390625,
      "learning_rate": 9.263157894736841e-07,
      "loss": 0.0,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 28
    },
    {
      "completion_length": 502.09375,
      "epoch": 0.7631578947368421,
      "grad_norm": 0.027645198038216884,
      "kl": 0.00115203857421875,
      "learning_rate": 9.236842105263157e-07,
      "loss": 0.0,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 29
    },
    {
      "completion_length": 512.3125,
      "epoch": 0.7894736842105263,
      "grad_norm": 0.04172906532475141,
      "kl": 0.001129150390625,
      "learning_rate": 9.210526315789473e-07,
      "loss": 0.0,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 30
    },
    {
      "completion_length": 511.8125,
      "epoch": 0.8157894736842105,
      "grad_norm": 1.4555609303743895,
      "kl": 0.00112152099609375,
      "learning_rate": 9.184210526315789e-07,
      "loss": 0.0,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 31
    },
    {
      "completion_length": 511.5,
      "epoch": 0.8421052631578947,
      "grad_norm": 0.9929122661974509,
      "kl": 0.00118255615234375,
      "learning_rate": 9.157894736842105e-07,
      "loss": 0.0,
      "reward": 1.875,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 0.96875,
      "step": 32
    },
    {
      "completion_length": 512.8125,
      "epoch": 0.868421052631579,
      "grad_norm": 0.6447662126840653,
      "kl": 0.0010833740234375,
      "learning_rate": 9.13157894736842e-07,
      "loss": 0.0,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.96875,
      "step": 33
    },
    {
      "completion_length": 490.28125,
      "epoch": 0.8947368421052632,
      "grad_norm": 0.03130511301152833,
      "kl": 0.001373291015625,
      "learning_rate": 9.105263157894737e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 34
    },
    {
      "completion_length": 492.34375,
      "epoch": 0.9210526315789473,
      "grad_norm": 0.03731473780437261,
      "kl": 0.001312255859375,
      "learning_rate": 9.078947368421053e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 35
    },
    {
      "completion_length": 500.09375,
      "epoch": 0.9473684210526315,
      "grad_norm": 0.8048521628683286,
      "kl": 0.001373291015625,
      "learning_rate": 9.052631578947368e-07,
      "loss": 0.0001,
      "reward": 1.9375,
      "reward_std": 0.07216878235340118,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 36
    },
    {
      "completion_length": 497.09375,
      "epoch": 0.9736842105263158,
      "grad_norm": 1.0588152232155654,
      "kl": 0.00146484375,
      "learning_rate": 9.026315789473684e-07,
      "loss": 0.0001,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 37
    },
    {
      "completion_length": 533.0,
      "epoch": 1.0,
      "grad_norm": 1.7019611225357512,
      "kl": 0.00144195556640625,
      "learning_rate": 9e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.10000000149011612,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 38
    },
    {
      "completion_length": 513.90625,
      "epoch": 1.0263157894736843,
      "grad_norm": 0.02315223199099485,
      "kl": 0.00136566162109375,
      "learning_rate": 8.973684210526315e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 39
    },
    {
      "completion_length": 484.03125,
      "epoch": 1.0526315789473684,
      "grad_norm": 0.5037354331914335,
      "kl": 0.00136566162109375,
      "learning_rate": 8.947368421052631e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 40
    },
    {
      "completion_length": 498.09375,
      "epoch": 1.0789473684210527,
      "grad_norm": 1.2181702218163735,
      "kl": 0.00150299072265625,
      "learning_rate": 8.921052631578947e-07,
      "loss": 0.0001,
      "reward": 1.78125,
      "reward_std": 0.13466878235340118,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.96875,
      "step": 41
    },
    {
      "completion_length": 520.96875,
      "epoch": 1.1052631578947367,
      "grad_norm": 3.2884258137508184,
      "kl": 0.001495361328125,
      "learning_rate": 8.894736842105263e-07,
      "loss": 0.0001,
      "reward": 1.9375,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 42
    },
    {
      "completion_length": 518.21875,
      "epoch": 1.131578947368421,
      "grad_norm": 1.762389574763672,
      "kl": 0.00171661376953125,
      "learning_rate": 8.868421052631579e-07,
      "loss": 0.0001,
      "reward": 1.875,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 43
    },
    {
      "completion_length": 514.15625,
      "epoch": 1.1578947368421053,
      "grad_norm": 0.027114452970964015,
      "kl": 0.00168609619140625,
      "learning_rate": 8.842105263157895e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 44
    },
    {
      "completion_length": 496.59375,
      "epoch": 1.1842105263157894,
      "grad_norm": 0.026564768246094845,
      "kl": 0.00182342529296875,
      "learning_rate": 8.815789473684209e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 45
    },
    {
      "completion_length": 504.6875,
      "epoch": 1.2105263157894737,
      "grad_norm": 1.8979762571267376,
      "kl": 0.0017852783203125,
      "learning_rate": 8.789473684210525e-07,
      "loss": 0.0001,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.96875,
      "step": 46
    },
    {
      "completion_length": 501.84375,
      "epoch": 1.236842105263158,
      "grad_norm": 0.03582040873644126,
      "kl": 0.0017242431640625,
      "learning_rate": 8.763157894736841e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 47
    },
    {
      "completion_length": 503.53125,
      "epoch": 1.263157894736842,
      "grad_norm": 2.144284016014784,
      "kl": 0.002044677734375,
      "learning_rate": 8.736842105263158e-07,
      "loss": 0.0001,
      "reward": 1.90625,
      "reward_std": 0.13466878235340118,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 48
    },
    {
      "completion_length": 504.90625,
      "epoch": 1.2894736842105263,
      "grad_norm": 0.03267048642715876,
      "kl": 0.001922607421875,
      "learning_rate": 8.710526315789474e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 49
    },
    {
      "completion_length": 488.125,
      "epoch": 1.3157894736842106,
      "grad_norm": 0.8667798461167672,
      "kl": 0.00182342529296875,
      "learning_rate": 8.684210526315789e-07,
      "loss": 0.0001,
      "reward": 1.9375,
      "reward_std": 0.07216878235340118,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 50
    },
    {
      "completion_length": 498.96875,
      "epoch": 1.3421052631578947,
      "grad_norm": 1.24790653131382,
      "kl": 0.0020599365234375,
      "learning_rate": 8.657894736842105e-07,
      "loss": 0.0001,
      "reward": 1.90625,
      "reward_std": 0.13466878235340118,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 51
    },
    {
      "completion_length": 519.3125,
      "epoch": 1.368421052631579,
      "grad_norm": 0.03953679886184328,
      "kl": 0.001983642578125,
      "learning_rate": 8.631578947368421e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 52
    },
    {
      "completion_length": 517.03125,
      "epoch": 1.3947368421052633,
      "grad_norm": 0.7933468259292563,
      "kl": 0.0022125244140625,
      "learning_rate": 8.605263157894737e-07,
      "loss": 0.0001,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 53
    },
    {
      "completion_length": 518.46875,
      "epoch": 1.4210526315789473,
      "grad_norm": 1.0167556818631809,
      "kl": 0.0021514892578125,
      "learning_rate": 8.578947368421053e-07,
      "loss": 0.0001,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 54
    },
    {
      "completion_length": 504.625,
      "epoch": 1.4473684210526316,
      "grad_norm": 0.030072954640137744,
      "kl": 0.00201416015625,
      "learning_rate": 8.552631578947367e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 55
    },
    {
      "completion_length": 487.09375,
      "epoch": 1.4736842105263157,
      "grad_norm": 0.044081948740351436,
      "kl": 0.00238037109375,
      "learning_rate": 8.526315789473683e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 56
    },
    {
      "completion_length": 494.84375,
      "epoch": 1.5,
      "grad_norm": 0.6659130960255057,
      "kl": 0.0023193359375,
      "learning_rate": 8.499999999999999e-07,
      "loss": 0.0001,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.96875,
      "step": 57
    },
    {
      "completion_length": 502.96875,
      "epoch": 1.526315789473684,
      "grad_norm": 1.2870163334442986,
      "kl": 0.0026397705078125,
      "learning_rate": 8.473684210526315e-07,
      "loss": 0.0001,
      "reward": 1.9375,
      "reward_std": 0.07216878235340118,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 58
    },
    {
      "completion_length": 509.15625,
      "epoch": 1.5526315789473686,
      "grad_norm": 0.768893645703431,
      "kl": 0.002044677734375,
      "learning_rate": 8.447368421052631e-07,
      "loss": 0.0001,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 59
    },
    {
      "completion_length": 498.71875,
      "epoch": 1.5789473684210527,
      "grad_norm": 1.2239445462190495,
      "kl": 0.002593994140625,
      "learning_rate": 8.421052631578947e-07,
      "loss": 0.0001,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.96875,
      "step": 60
    },
    {
      "completion_length": 519.40625,
      "epoch": 1.6052631578947367,
      "grad_norm": 2.677035126267462,
      "kl": 0.0023193359375,
      "learning_rate": 8.394736842105262e-07,
      "loss": 0.0001,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.96875,
      "step": 61
    },
    {
      "completion_length": 505.1875,
      "epoch": 1.631578947368421,
      "grad_norm": 1.8844544141537973,
      "kl": 0.002593994140625,
      "learning_rate": 8.368421052631579e-07,
      "loss": 0.0001,
      "reward": 1.90625,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 62
    },
    {
      "completion_length": 498.59375,
      "epoch": 1.6578947368421053,
      "grad_norm": 1.65261105976823,
      "kl": 0.00244140625,
      "learning_rate": 8.342105263157895e-07,
      "loss": 0.0001,
      "reward": 1.90625,
      "reward_std": 0.13466878235340118,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 63
    },
    {
      "completion_length": 517.0,
      "epoch": 1.6842105263157894,
      "grad_norm": 1.2268440042399227,
      "kl": 0.002655029296875,
      "learning_rate": 8.315789473684211e-07,
      "loss": 0.0001,
      "reward": 1.9375,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 64
    },
    {
      "completion_length": 489.53125,
      "epoch": 1.7105263157894737,
      "grad_norm": 1.1558248277077878,
      "kl": 0.0027008056640625,
      "learning_rate": 8.289473684210527e-07,
      "loss": 0.0001,
      "reward": 1.90625,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 65
    },
    {
      "completion_length": 501.09375,
      "epoch": 1.736842105263158,
      "grad_norm": 0.02786918187085701,
      "kl": 0.002227783203125,
      "learning_rate": 8.263157894736841e-07,
      "loss": 0.0001,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 66
    },
    {
      "completion_length": 512.65625,
      "epoch": 1.763157894736842,
      "grad_norm": 1.4661151060918443,
      "kl": 0.0027313232421875,
      "learning_rate": 8.236842105263157e-07,
      "loss": 0.0001,
      "reward": 1.9375,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 0.96875,
      "step": 67
    },
    {
      "completion_length": 516.90625,
      "epoch": 1.7894736842105263,
      "grad_norm": 0.03947664659715401,
      "kl": 0.0025482177734375,
      "learning_rate": 8.210526315789473e-07,
      "loss": 0.0001,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 68
    },
    {
      "completion_length": 534.34375,
      "epoch": 1.8157894736842106,
      "grad_norm": 0.0321644121580206,
      "kl": 0.002655029296875,
      "learning_rate": 8.184210526315789e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 69
    },
    {
      "completion_length": 497.1875,
      "epoch": 1.8421052631578947,
      "grad_norm": 0.8511530790848818,
      "kl": 0.002960205078125,
      "learning_rate": 8.157894736842105e-07,
      "loss": 0.0001,
      "reward": 1.9375,
      "reward_std": 0.07216878235340118,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 70
    },
    {
      "completion_length": 504.625,
      "epoch": 1.868421052631579,
      "grad_norm": 0.045403243082258384,
      "kl": 0.0028228759765625,
      "learning_rate": 8.131578947368421e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 71
    },
    {
      "completion_length": 500.15625,
      "epoch": 1.8947368421052633,
      "grad_norm": 1.7186629257319292,
      "kl": 0.0026397705078125,
      "learning_rate": 8.105263157894736e-07,
      "loss": 0.0001,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.96875,
      "step": 72
    },
    {
      "completion_length": 502.96875,
      "epoch": 1.9210526315789473,
      "grad_norm": 0.04829244770549609,
      "kl": 0.00244140625,
      "learning_rate": 8.078947368421052e-07,
      "loss": 0.0001,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 73
    },
    {
      "completion_length": 503.3125,
      "epoch": 1.9473684210526314,
      "grad_norm": 0.8916174408258744,
      "kl": 0.0023345947265625,
      "learning_rate": 8.052631578947368e-07,
      "loss": 0.0001,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 74
    },
    {
      "completion_length": 494.625,
      "epoch": 1.973684210526316,
      "grad_norm": 1.1293143521918012,
      "kl": 0.002655029296875,
      "learning_rate": 8.026315789473685e-07,
      "loss": 0.0001,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 75
    },
    {
      "completion_length": 500.0,
      "epoch": 2.0,
      "grad_norm": 3.37331916516421,
      "kl": 0.0027923583984375,
      "learning_rate": 8e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 76
    },
    {
      "completion_length": 531.96875,
      "epoch": 2.026315789473684,
      "grad_norm": 0.030167855476728275,
      "kl": 0.0027618408203125,
      "learning_rate": 7.973684210526315e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 77
    },
    {
      "completion_length": 499.34375,
      "epoch": 2.0526315789473686,
      "grad_norm": 1.501743158573559,
      "kl": 0.00311279296875,
      "learning_rate": 7.947368421052631e-07,
      "loss": 0.0001,
      "reward": 1.8125,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 0.96875,
      "step": 78
    },
    {
      "completion_length": 492.78125,
      "epoch": 2.0789473684210527,
      "grad_norm": 0.03689606923255948,
      "kl": 0.002838134765625,
      "learning_rate": 7.921052631578947e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 79
    },
    {
      "completion_length": 528.5,
      "epoch": 2.1052631578947367,
      "grad_norm": 1.3204028049443157,
      "kl": 0.0029144287109375,
      "learning_rate": 7.894736842105263e-07,
      "loss": 0.0001,
      "reward": 1.90625,
      "reward_std": 0.1875,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 0.96875,
      "step": 80
    },
    {
      "completion_length": 513.8125,
      "epoch": 2.1315789473684212,
      "grad_norm": 0.04731343670470243,
      "kl": 0.00262451171875,
      "learning_rate": 7.868421052631579e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 81
    },
    {
      "completion_length": 500.4375,
      "epoch": 2.1578947368421053,
      "grad_norm": 0.7278429400242111,
      "kl": 0.00262451171875,
      "learning_rate": 7.842105263157895e-07,
      "loss": 0.0001,
      "reward": 1.90625,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 82
    },
    {
      "completion_length": 520.71875,
      "epoch": 2.1842105263157894,
      "grad_norm": 0.9942768526326492,
      "kl": 0.00274658203125,
      "learning_rate": 7.81578947368421e-07,
      "loss": 0.0001,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 83
    },
    {
      "completion_length": 529.3125,
      "epoch": 2.2105263157894735,
      "grad_norm": 0.028533047509329622,
      "kl": 0.00244140625,
      "learning_rate": 7.789473684210526e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 84
    },
    {
      "completion_length": 520.78125,
      "epoch": 2.236842105263158,
      "grad_norm": 1.454115482202384,
      "kl": 0.002838134765625,
      "learning_rate": 7.763157894736841e-07,
      "loss": 0.0001,
      "reward": 1.875,
      "reward_std": 0.14433756470680237,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 85
    },
    {
      "completion_length": 493.6875,
      "epoch": 2.263157894736842,
      "grad_norm": 0.034131542034081065,
      "kl": 0.002899169921875,
      "learning_rate": 7.736842105263157e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 86
    },
    {
      "completion_length": 495.46875,
      "epoch": 2.2894736842105265,
      "grad_norm": 0.04357390543835616,
      "kl": 0.0027313232421875,
      "learning_rate": 7.710526315789473e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 87
    },
    {
      "completion_length": 508.875,
      "epoch": 2.3157894736842106,
      "grad_norm": 1.0055553420607368,
      "kl": 0.0022735595703125,
      "learning_rate": 7.684210526315788e-07,
      "loss": 0.0001,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 88
    },
    {
      "completion_length": 497.6875,
      "epoch": 2.3421052631578947,
      "grad_norm": 1.6914903798873342,
      "kl": 0.0027313232421875,
      "learning_rate": 7.657894736842105e-07,
      "loss": 0.0001,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 89
    },
    {
      "completion_length": 512.25,
      "epoch": 2.3684210526315788,
      "grad_norm": 0.02834284445323136,
      "kl": 0.002655029296875,
      "learning_rate": 7.631578947368421e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 90
    },
    {
      "completion_length": 522.875,
      "epoch": 2.3947368421052633,
      "grad_norm": 0.873621683810661,
      "kl": 0.0032806396484375,
      "learning_rate": 7.605263157894737e-07,
      "loss": 0.0001,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 91
    },
    {
      "completion_length": 502.59375,
      "epoch": 2.4210526315789473,
      "grad_norm": 0.038730383952057044,
      "kl": 0.0035247802734375,
      "learning_rate": 7.578947368421053e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 92
    },
    {
      "completion_length": 522.78125,
      "epoch": 2.4473684210526314,
      "grad_norm": 1.405567089260259,
      "kl": 0.0030670166015625,
      "learning_rate": 7.552631578947369e-07,
      "loss": 0.0001,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 93
    },
    {
      "completion_length": 505.46875,
      "epoch": 2.473684210526316,
      "grad_norm": 0.030020529643763186,
      "kl": 0.0025634765625,
      "learning_rate": 7.526315789473684e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 94
    },
    {
      "completion_length": 510.5625,
      "epoch": 2.5,
      "grad_norm": 1.1924545507060844,
      "kl": 0.003173828125,
      "learning_rate": 7.5e-07,
      "loss": 0.0001,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 95
    },
    {
      "completion_length": 522.15625,
      "epoch": 2.526315789473684,
      "grad_norm": 0.043988683847166775,
      "kl": 0.003204345703125,
      "learning_rate": 7.473684210526315e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 96
    },
    {
      "completion_length": 506.21875,
      "epoch": 2.5526315789473686,
      "grad_norm": 1.2599814660966382,
      "kl": 0.00372314453125,
      "learning_rate": 7.447368421052631e-07,
      "loss": 0.0001,
      "reward": 1.90625,
      "reward_std": 0.13466878235340118,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 0.96875,
      "step": 97
    },
    {
      "completion_length": 511.0625,
      "epoch": 2.5789473684210527,
      "grad_norm": 0.042162872448447394,
      "kl": 0.0033111572265625,
      "learning_rate": 7.421052631578947e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 98
    },
    {
      "completion_length": 504.90625,
      "epoch": 2.6052631578947367,
      "grad_norm": 0.0341121471676267,
      "kl": 0.0031280517578125,
      "learning_rate": 7.394736842105262e-07,
      "loss": 0.0001,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 99
    },
    {
      "completion_length": 501.53125,
      "epoch": 2.6315789473684212,
      "grad_norm": 2.566911802127545,
      "kl": 0.003265380859375,
      "learning_rate": 7.368421052631578e-07,
      "loss": 0.0001,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 100
    },
    {
      "completion_length": 498.875,
      "epoch": 2.6578947368421053,
      "grad_norm": 0.03337527256197859,
      "kl": 0.003082275390625,
      "learning_rate": 7.342105263157894e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 101
    },
    {
      "completion_length": 497.65625,
      "epoch": 2.6842105263157894,
      "grad_norm": 0.04741778746371284,
      "kl": 0.003662109375,
      "learning_rate": 7.315789473684211e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 102
    },
    {
      "completion_length": 487.15625,
      "epoch": 2.7105263157894735,
      "grad_norm": 1.8393077262931932,
      "kl": 0.00372314453125,
      "learning_rate": 7.289473684210527e-07,
      "loss": 0.0001,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 103
    },
    {
      "completion_length": 521.0625,
      "epoch": 2.736842105263158,
      "grad_norm": 0.03889186679405788,
      "kl": 0.0036468505859375,
      "learning_rate": 7.263157894736843e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 104
    },
    {
      "completion_length": 510.5625,
      "epoch": 2.763157894736842,
      "grad_norm": 1.447109219593432,
      "kl": 0.0037078857421875,
      "learning_rate": 7.236842105263158e-07,
      "loss": 0.0001,
      "reward": 1.8125,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 0.96875,
      "step": 105
    },
    {
      "completion_length": 509.71875,
      "epoch": 2.7894736842105265,
      "grad_norm": 1.4858428116278655,
      "kl": 0.003326416015625,
      "learning_rate": 7.210526315789473e-07,
      "loss": 0.0001,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 106
    },
    {
      "completion_length": 508.40625,
      "epoch": 2.8157894736842106,
      "grad_norm": 0.961549949503767,
      "kl": 0.00323486328125,
      "learning_rate": 7.184210526315789e-07,
      "loss": 0.0001,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 107
    },
    {
      "completion_length": 510.0625,
      "epoch": 2.8421052631578947,
      "grad_norm": 1.202043926324514,
      "kl": 0.0030670166015625,
      "learning_rate": 7.157894736842105e-07,
      "loss": 0.0001,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 108
    },
    {
      "completion_length": 501.46875,
      "epoch": 2.8684210526315788,
      "grad_norm": 0.8258294441704436,
      "kl": 0.0034637451171875,
      "learning_rate": 7.131578947368421e-07,
      "loss": 0.0001,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.96875,
      "step": 109
    },
    {
      "completion_length": 536.15625,
      "epoch": 2.8947368421052633,
      "grad_norm": 0.7881046328867846,
      "kl": 0.003448486328125,
      "learning_rate": 7.105263157894736e-07,
      "loss": 0.0001,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 110
    },
    {
      "completion_length": 506.75,
      "epoch": 2.9210526315789473,
      "grad_norm": 0.04116293609934439,
      "kl": 0.0035858154296875,
      "learning_rate": 7.078947368421052e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 111
    },
    {
      "completion_length": 521.1875,
      "epoch": 2.9473684210526314,
      "grad_norm": 1.0874707114277227,
      "kl": 0.0035552978515625,
      "learning_rate": 7.052631578947368e-07,
      "loss": 0.0001,
      "reward": 1.78125,
      "reward_std": 0.13466878235340118,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 112
    },
    {
      "completion_length": 511.28125,
      "epoch": 2.973684210526316,
      "grad_norm": 0.0625092599479988,
      "kl": 0.003448486328125,
      "learning_rate": 7.026315789473684e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 113
    },
    {
      "completion_length": 503.8000183105469,
      "epoch": 3.0,
      "grad_norm": 0.03814663470015353,
      "kl": 0.0034332275390625,
      "learning_rate": 7e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 114
    },
    {
      "completion_length": 507.78125,
      "epoch": 3.026315789473684,
      "grad_norm": 0.05686010982708928,
      "kl": 0.0034027099609375,
      "learning_rate": 6.973684210526314e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 115
    },
    {
      "completion_length": 536.5,
      "epoch": 3.0526315789473686,
      "grad_norm": 0.04279707484724643,
      "kl": 0.003021240234375,
      "learning_rate": 6.947368421052631e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 116
    },
    {
      "completion_length": 504.84375,
      "epoch": 3.0789473684210527,
      "grad_norm": 0.8334012538858381,
      "kl": 0.0032958984375,
      "learning_rate": 6.921052631578947e-07,
      "loss": 0.0001,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 117
    },
    {
      "completion_length": 499.84375,
      "epoch": 3.1052631578947367,
      "grad_norm": 0.9953068967016894,
      "kl": 0.0040283203125,
      "learning_rate": 6.894736842105263e-07,
      "loss": 0.0002,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 118
    },
    {
      "completion_length": 511.59375,
      "epoch": 3.1315789473684212,
      "grad_norm": 0.03618191894604707,
      "kl": 0.0034027099609375,
      "learning_rate": 6.868421052631579e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 119
    },
    {
      "completion_length": 521.28125,
      "epoch": 3.1578947368421053,
      "grad_norm": 1.3499675363862598,
      "kl": 0.00384521484375,
      "learning_rate": 6.842105263157895e-07,
      "loss": 0.0002,
      "reward": 1.9375,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 120
    },
    {
      "completion_length": 508.84375,
      "epoch": 3.1842105263157894,
      "grad_norm": 1.3905859323416692,
      "kl": 0.0034942626953125,
      "learning_rate": 6.81578947368421e-07,
      "loss": 0.0001,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 121
    },
    {
      "completion_length": 517.40625,
      "epoch": 3.2105263157894735,
      "grad_norm": 1.3622360911755893,
      "kl": 0.0035552978515625,
      "learning_rate": 6.789473684210526e-07,
      "loss": 0.0001,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 122
    },
    {
      "completion_length": 495.875,
      "epoch": 3.236842105263158,
      "grad_norm": 1.451454972303075,
      "kl": 0.003509521484375,
      "learning_rate": 6.763157894736842e-07,
      "loss": 0.0001,
      "reward": 1.875,
      "reward_std": 0.19716878235340118,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 0.96875,
      "step": 123
    },
    {
      "completion_length": 498.40625,
      "epoch": 3.263157894736842,
      "grad_norm": 0.06927241278752044,
      "kl": 0.0036773681640625,
      "learning_rate": 6.736842105263158e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 124
    },
    {
      "completion_length": 518.03125,
      "epoch": 3.2894736842105265,
      "grad_norm": 0.8762828465082988,
      "kl": 0.0033721923828125,
      "learning_rate": 6.710526315789473e-07,
      "loss": 0.0001,
      "reward": 1.875,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 0.96875,
      "step": 125
    },
    {
      "completion_length": 525.96875,
      "epoch": 3.3157894736842106,
      "grad_norm": 0.1279472603343596,
      "kl": 0.0032196044921875,
      "learning_rate": 6.684210526315788e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 126
    },
    {
      "completion_length": 488.0625,
      "epoch": 3.3421052631578947,
      "grad_norm": 0.31566337930874167,
      "kl": 0.0032196044921875,
      "learning_rate": 6.657894736842104e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 127
    },
    {
      "completion_length": 513.09375,
      "epoch": 3.3684210526315788,
      "grad_norm": 2.827017411958231,
      "kl": 0.00457763671875,
      "learning_rate": 6.63157894736842e-07,
      "loss": 0.0002,
      "reward": 1.8125,
      "reward_std": 0.25,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 128
    },
    {
      "completion_length": 494.71875,
      "epoch": 3.3947368421052633,
      "grad_norm": 1.1860759612828082,
      "kl": 0.003631591796875,
      "learning_rate": 6.605263157894737e-07,
      "loss": 0.0001,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 129
    },
    {
      "completion_length": 512.53125,
      "epoch": 3.4210526315789473,
      "grad_norm": 0.03855778832036919,
      "kl": 0.0036468505859375,
      "learning_rate": 6.578947368421053e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 130
    },
    {
      "completion_length": 519.5625,
      "epoch": 3.4473684210526314,
      "grad_norm": 0.04402464100179393,
      "kl": 0.00347900390625,
      "learning_rate": 6.552631578947369e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 131
    },
    {
      "completion_length": 486.21875,
      "epoch": 3.473684210526316,
      "grad_norm": 0.03890279645364714,
      "kl": 0.003509521484375,
      "learning_rate": 6.526315789473684e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 132
    },
    {
      "completion_length": 497.28125,
      "epoch": 3.5,
      "grad_norm": 0.9474366715612303,
      "kl": 0.004364013671875,
      "learning_rate": 6.5e-07,
      "loss": 0.0002,
      "reward": 1.90625,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 133
    },
    {
      "completion_length": 495.59375,
      "epoch": 3.526315789473684,
      "grad_norm": 1.3876819486259453,
      "kl": 0.0037689208984375,
      "learning_rate": 6.473684210526316e-07,
      "loss": 0.0002,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 134
    },
    {
      "completion_length": 509.0,
      "epoch": 3.5526315789473686,
      "grad_norm": 0.04968025093227853,
      "kl": 0.0034027099609375,
      "learning_rate": 6.447368421052632e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 135
    },
    {
      "completion_length": 512.90625,
      "epoch": 3.5789473684210527,
      "grad_norm": 2.6370590276884385,
      "kl": 0.00433349609375,
      "learning_rate": 6.421052631578947e-07,
      "loss": 0.0002,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.96875,
      "step": 136
    },
    {
      "completion_length": 515.46875,
      "epoch": 3.6052631578947367,
      "grad_norm": 0.9919508500122917,
      "kl": 0.00469970703125,
      "learning_rate": 6.394736842105262e-07,
      "loss": 0.0002,
      "reward": 1.9375,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 137
    },
    {
      "completion_length": 486.75,
      "epoch": 3.6315789473684212,
      "grad_norm": 0.03323526791149113,
      "kl": 0.003204345703125,
      "learning_rate": 6.368421052631578e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 138
    },
    {
      "completion_length": 503.5625,
      "epoch": 3.6578947368421053,
      "grad_norm": 1.1252637437400366,
      "kl": 0.0037841796875,
      "learning_rate": 6.342105263157894e-07,
      "loss": 0.0002,
      "reward": 1.90625,
      "reward_std": 0.13466878235340118,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 139
    },
    {
      "completion_length": 517.40625,
      "epoch": 3.6842105263157894,
      "grad_norm": 0.8025074781730501,
      "kl": 0.004150390625,
      "learning_rate": 6.31578947368421e-07,
      "loss": 0.0002,
      "reward": 1.9375,
      "reward_std": 0.07216878235340118,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 140
    },
    {
      "completion_length": 510.21875,
      "epoch": 3.7105263157894735,
      "grad_norm": 0.032038711480624,
      "kl": 0.003448486328125,
      "learning_rate": 6.289473684210526e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 141
    },
    {
      "completion_length": 505.65625,
      "epoch": 3.736842105263158,
      "grad_norm": 0.030986940813949752,
      "kl": 0.0034637451171875,
      "learning_rate": 6.263157894736842e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 142
    },
    {
      "completion_length": 521.4375,
      "epoch": 3.763157894736842,
      "grad_norm": 0.03486593372780562,
      "kl": 0.00341796875,
      "learning_rate": 6.236842105263158e-07,
      "loss": 0.0001,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 143
    },
    {
      "completion_length": 503.1875,
      "epoch": 3.7894736842105265,
      "grad_norm": 0.9150417772985632,
      "kl": 0.004150390625,
      "learning_rate": 6.210526315789474e-07,
      "loss": 0.0002,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.96875,
      "step": 144
    },
    {
      "completion_length": 504.9375,
      "epoch": 3.8157894736842106,
      "grad_norm": 1.2738344698727906,
      "kl": 0.00396728515625,
      "learning_rate": 6.18421052631579e-07,
      "loss": 0.0002,
      "reward": 1.90625,
      "reward_std": 0.13466878235340118,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 145
    },
    {
      "completion_length": 503.25,
      "epoch": 3.8421052631578947,
      "grad_norm": 1.1946263189029565,
      "kl": 0.0034332275390625,
      "learning_rate": 6.157894736842105e-07,
      "loss": 0.0001,
      "reward": 1.875,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 146
    },
    {
      "completion_length": 534.65625,
      "epoch": 3.8684210526315788,
      "grad_norm": 1.0916070617134428,
      "kl": 0.003570556640625,
      "learning_rate": 6.131578947368421e-07,
      "loss": 0.0001,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 147
    },
    {
      "completion_length": 496.53125,
      "epoch": 3.8947368421052633,
      "grad_norm": 0.04197200000816943,
      "kl": 0.00335693359375,
      "learning_rate": 6.105263157894736e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 148
    },
    {
      "completion_length": 505.0625,
      "epoch": 3.9210526315789473,
      "grad_norm": 0.9112054136826518,
      "kl": 0.0034942626953125,
      "learning_rate": 6.078947368421052e-07,
      "loss": 0.0001,
      "reward": 1.9375,
      "reward_std": 0.07216878235340118,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 149
    },
    {
      "completion_length": 491.9375,
      "epoch": 3.9473684210526314,
      "grad_norm": 0.896485459132187,
      "kl": 0.0036468505859375,
      "learning_rate": 6.052631578947368e-07,
      "loss": 0.0001,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 150
    },
    {
      "completion_length": 497.46875,
      "epoch": 3.973684210526316,
      "grad_norm": 0.7253680478752668,
      "kl": 0.004150390625,
      "learning_rate": 6.026315789473684e-07,
      "loss": 0.0002,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 151
    },
    {
      "completion_length": 455.3999938964844,
      "epoch": 4.0,
      "grad_norm": 0.04315556382921662,
      "kl": 0.003662109375,
      "learning_rate": 6e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 152
    },
    {
      "completion_length": 503.625,
      "epoch": 4.026315789473684,
      "grad_norm": 0.03450640103445653,
      "kl": 0.0038299560546875,
      "learning_rate": 5.973684210526316e-07,
      "loss": 0.0002,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 153
    },
    {
      "completion_length": 510.9375,
      "epoch": 4.052631578947368,
      "grad_norm": 0.8027374171690022,
      "kl": 0.003326416015625,
      "learning_rate": 5.947368421052631e-07,
      "loss": 0.0001,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 154
    },
    {
      "completion_length": 528.375,
      "epoch": 4.078947368421052,
      "grad_norm": 0.834678814594713,
      "kl": 0.0042724609375,
      "learning_rate": 5.921052631578946e-07,
      "loss": 0.0002,
      "reward": 1.9375,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 0.96875,
      "step": 155
    },
    {
      "completion_length": 528.59375,
      "epoch": 4.105263157894737,
      "grad_norm": 1.5126161433832532,
      "kl": 0.0036468505859375,
      "learning_rate": 5.894736842105262e-07,
      "loss": 0.0001,
      "reward": 1.90625,
      "reward_std": 0.13466878235340118,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 156
    },
    {
      "completion_length": 499.21875,
      "epoch": 4.131578947368421,
      "grad_norm": 0.04002314203603394,
      "kl": 0.003570556640625,
      "learning_rate": 5.868421052631579e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 157
    },
    {
      "completion_length": 517.53125,
      "epoch": 4.157894736842105,
      "grad_norm": 0.8871107366636743,
      "kl": 0.004241943359375,
      "learning_rate": 5.842105263157895e-07,
      "loss": 0.0002,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 158
    },
    {
      "completion_length": 500.46875,
      "epoch": 4.184210526315789,
      "grad_norm": 0.034607701527423755,
      "kl": 0.0036773681640625,
      "learning_rate": 5.81578947368421e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 159
    },
    {
      "completion_length": 508.25,
      "epoch": 4.2105263157894735,
      "grad_norm": 0.8214538935630136,
      "kl": 0.00439453125,
      "learning_rate": 5.789473684210526e-07,
      "loss": 0.0002,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 160
    },
    {
      "completion_length": 510.375,
      "epoch": 4.2368421052631575,
      "grad_norm": 1.6395530042768915,
      "kl": 0.00408935546875,
      "learning_rate": 5.763157894736842e-07,
      "loss": 0.0002,
      "reward": 1.84375,
      "reward_std": 0.1875,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 161
    },
    {
      "completion_length": 514.75,
      "epoch": 4.2631578947368425,
      "grad_norm": 2.6138224629827476,
      "kl": 0.0037841796875,
      "learning_rate": 5.736842105263158e-07,
      "loss": 0.0002,
      "reward": 1.90625,
      "reward_std": 0.13466878235340118,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 162
    },
    {
      "completion_length": 499.09375,
      "epoch": 4.2894736842105265,
      "grad_norm": 0.04808481003397733,
      "kl": 0.003997802734375,
      "learning_rate": 5.710526315789474e-07,
      "loss": 0.0002,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 163
    },
    {
      "completion_length": 506.03125,
      "epoch": 4.315789473684211,
      "grad_norm": 0.08804071935459232,
      "kl": 0.0038909912109375,
      "learning_rate": 5.68421052631579e-07,
      "loss": 0.0002,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 164
    },
    {
      "completion_length": 476.03125,
      "epoch": 4.342105263157895,
      "grad_norm": 0.03572194363283224,
      "kl": 0.0036773681640625,
      "learning_rate": 5.657894736842104e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 165
    },
    {
      "completion_length": 536.6875,
      "epoch": 4.368421052631579,
      "grad_norm": 0.04019933989542527,
      "kl": 0.0038909912109375,
      "learning_rate": 5.63157894736842e-07,
      "loss": 0.0002,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 166
    },
    {
      "completion_length": 509.25,
      "epoch": 4.394736842105263,
      "grad_norm": 1.5681592918937646,
      "kl": 0.005035400390625,
      "learning_rate": 5.605263157894736e-07,
      "loss": 0.0002,
      "reward": 1.84375,
      "reward_std": 0.1875,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 167
    },
    {
      "completion_length": 497.34375,
      "epoch": 4.421052631578947,
      "grad_norm": 2.538787505539558,
      "kl": 0.0042724609375,
      "learning_rate": 5.578947368421052e-07,
      "loss": 0.0002,
      "reward": 1.9375,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 168
    },
    {
      "completion_length": 520.65625,
      "epoch": 4.447368421052632,
      "grad_norm": 0.041019565179391565,
      "kl": 0.0032196044921875,
      "learning_rate": 5.552631578947368e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 169
    },
    {
      "completion_length": 511.71875,
      "epoch": 4.473684210526316,
      "grad_norm": 1.0243322126591785,
      "kl": 0.00445556640625,
      "learning_rate": 5.526315789473684e-07,
      "loss": 0.0002,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 170
    },
    {
      "completion_length": 523.40625,
      "epoch": 4.5,
      "grad_norm": 0.6764116645829102,
      "kl": 0.00433349609375,
      "learning_rate": 5.5e-07,
      "loss": 0.0002,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.96875,
      "step": 171
    },
    {
      "completion_length": 501.75,
      "epoch": 4.526315789473684,
      "grad_norm": 1.0831256376423706,
      "kl": 0.003814697265625,
      "learning_rate": 5.473684210526316e-07,
      "loss": 0.0002,
      "reward": 1.875,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 172
    },
    {
      "completion_length": 506.21875,
      "epoch": 4.552631578947368,
      "grad_norm": 0.046752954659444006,
      "kl": 0.00360107421875,
      "learning_rate": 5.447368421052632e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 173
    },
    {
      "completion_length": 497.03125,
      "epoch": 4.578947368421053,
      "grad_norm": 1.0272776092127476,
      "kl": 0.0034637451171875,
      "learning_rate": 5.421052631578948e-07,
      "loss": 0.0001,
      "reward": 1.90625,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 174
    },
    {
      "completion_length": 498.40625,
      "epoch": 4.605263157894737,
      "grad_norm": 1.0347251781483067,
      "kl": 0.003662109375,
      "learning_rate": 5.394736842105264e-07,
      "loss": 0.0001,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 175
    },
    {
      "completion_length": 492.0625,
      "epoch": 4.631578947368421,
      "grad_norm": 5.431342136617613,
      "kl": 0.004364013671875,
      "learning_rate": 5.368421052631578e-07,
      "loss": 0.0002,
      "reward": 1.9375,
      "reward_std": 0.07216878235340118,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 176
    },
    {
      "completion_length": 500.21875,
      "epoch": 4.657894736842105,
      "grad_norm": 0.061596768702879764,
      "kl": 0.003631591796875,
      "learning_rate": 5.342105263157894e-07,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 177
    },
    {
      "completion_length": 488.40625,
      "epoch": 4.684210526315789,
      "grad_norm": 0.03793363317517718,
      "kl": 0.004058837890625,
      "learning_rate": 5.31578947368421e-07,
      "loss": 0.0002,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 178
    },
    {
      "completion_length": 488.125,
      "epoch": 4.7105263157894735,
      "grad_norm": 1.459873859774728,
      "kl": 0.00433349609375,
      "learning_rate": 5.289473684210526e-07,
      "loss": 0.0002,
      "reward": 1.90625,
      "reward_std": 0.1875,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 179
    },
    {
      "completion_length": 522.0,
      "epoch": 4.7368421052631575,
      "grad_norm": 0.8661182876644632,
      "kl": 0.004638671875,
      "learning_rate": 5.263157894736842e-07,
      "loss": 0.0002,
      "reward": 1.9375,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 0.96875,
      "step": 180
    },
    {
      "completion_length": 502.6875,
      "epoch": 4.7631578947368425,
      "grad_norm": 0.07996281170258902,
      "kl": 0.004486083984375,
      "learning_rate": 5.236842105263157e-07,
      "loss": 0.0002,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 181
    },
    {
      "completion_length": 495.8125,
      "epoch": 4.7894736842105265,
      "grad_norm": 0.9651399175233428,
      "kl": 0.00360107421875,
      "learning_rate": 5.210526315789473e-07,
      "loss": 0.0001,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.96875,
      "step": 182
    },
    {
      "completion_length": 490.78125,
      "epoch": 4.815789473684211,
      "grad_norm": 0.957605365453321,
      "kl": 0.00421142578125,
      "learning_rate": 5.184210526315789e-07,
      "loss": 0.0002,
      "reward": 1.90625,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 183
    },
    {
      "completion_length": 489.53125,
      "epoch": 4.842105263157895,
      "grad_norm": 0.04438711227285697,
      "kl": 0.004974365234375,
      "learning_rate": 5.157894736842106e-07,
      "loss": 0.0002,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 184
    },
    {
      "completion_length": 491.125,
      "epoch": 4.868421052631579,
      "grad_norm": 0.7517932097620719,
      "kl": 0.00372314453125,
      "learning_rate": 5.131578947368422e-07,
      "loss": 0.0001,
      "reward": 1.9375,
      "reward_std": 0.07216878235340118,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 185
    },
    {
      "completion_length": 493.3125,
      "epoch": 4.894736842105263,
      "grad_norm": 1.5201290264209482,
      "kl": 0.005126953125,
      "learning_rate": 5.105263157894736e-07,
      "loss": 0.0002,
      "reward": 1.96875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 186
    },
    {
      "completion_length": 514.46875,
      "epoch": 4.921052631578947,
      "grad_norm": 0.0748706505594432,
      "kl": 0.004302978515625,
      "learning_rate": 5.078947368421052e-07,
      "loss": 0.0002,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 187
    },
    {
      "completion_length": 502.59375,
      "epoch": 4.947368421052632,
      "grad_norm": 1.0868228477221615,
      "kl": 0.004425048828125,
      "learning_rate": 5.052631578947368e-07,
      "loss": 0.0002,
      "reward": 1.90625,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 188
    },
    {
      "completion_length": 510.46875,
      "epoch": 4.973684210526316,
      "grad_norm": 0.03543275147216106,
      "kl": 0.0042724609375,
      "learning_rate": 5.026315789473684e-07,
      "loss": 0.0002,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 189
    },
    {
      "completion_length": 515.0,
      "epoch": 5.0,
      "grad_norm": 1.2489724519707108,
      "kl": 0.0037078857421875,
      "learning_rate": 5e-07,
      "loss": 0.0002,
      "reward": 1.8000000715255737,
      "reward_std": 0.4000000059604645,
      "rewards/accuracy_reward": 0.800000011920929,
      "rewards/format_reward": 1.0,
      "step": 190
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 380,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 38,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}