{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.49875311720698257,
  "eval_steps": 500,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 53.11458396911621,
      "epoch": 0.004987531172069825,
      "grad_norm": 2.589242696762085,
      "kl": 0.0,
      "learning_rate": 1e-06,
      "loss": -0.0,
      "reward": 1.84375,
      "reward_std": 0.5593957304954529,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.5625000298023224,
      "rewards/value_reward": 0.2812500074505806,
      "step": 1
    },
    {
      "completion_length": 54.47916793823242,
      "epoch": 0.00997506234413965,
      "grad_norm": 4.696207523345947,
      "kl": 0.00023698806762695312,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 1.6979166865348816,
      "reward_std": 0.47182925045490265,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.5208333432674408,
      "rewards/value_reward": 0.1770833395421505,
      "step": 2
    },
    {
      "completion_length": 58.270835876464844,
      "epoch": 0.014962593516209476,
      "grad_norm": 6.94147253036499,
      "kl": 0.00024080276489257812,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 1.6145833730697632,
      "reward_std": 0.38977013528347015,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.5104166865348816,
      "rewards/value_reward": 0.1041666716337204,
      "step": 3
    },
    {
      "completion_length": 43.98958396911621,
      "epoch": 0.0199501246882793,
      "grad_norm": 2.8042056560516357,
      "kl": 0.0001938343048095703,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 1.7187501192092896,
      "reward_std": 0.5527614653110504,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.53125,
      "rewards/value_reward": 0.1875000074505806,
      "step": 4
    },
    {
      "completion_length": 58.86458396911621,
      "epoch": 0.02493765586034913,
      "grad_norm": 6.230152130126953,
      "kl": 0.00038242340087890625,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 1.9895833730697632,
      "reward_std": 0.5839709639549255,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.7395833432674408,
      "rewards/value_reward": 0.25,
      "step": 5
    },
    {
      "completion_length": 50.95833396911621,
      "epoch": 0.029925187032418952,
      "grad_norm": 5.232593536376953,
      "kl": 0.0010242462158203125,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 1.6562500596046448,
      "reward_std": 0.5727283954620361,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.5000000298023224,
      "rewards/value_reward": 0.1562500074505806,
      "step": 6
    },
    {
      "completion_length": 55.6875,
      "epoch": 0.034912718204488775,
      "grad_norm": 4.557653427124023,
      "kl": 0.000457763671875,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 1.6041666865348816,
      "reward_std": 0.3993261456489563,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.5312500149011612,
      "rewards/value_reward": 0.0729166679084301,
      "step": 7
    },
    {
      "completion_length": 53.14583396911621,
      "epoch": 0.0399002493765586,
      "grad_norm": 6.5244011878967285,
      "kl": 0.0004901885986328125,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 1.8125000596046448,
      "reward_std": 0.511884331703186,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.6458333730697632,
      "rewards/value_reward": 0.1666666716337204,
      "step": 8
    },
    {
      "completion_length": 52.87500190734863,
      "epoch": 0.04488778054862843,
      "grad_norm": 3.019259452819824,
      "kl": 0.0007963180541992188,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 1.5000000596046448,
      "reward_std": 0.5527279376983643,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.4062500149011612,
      "rewards/value_reward": 0.0937500037252903,
      "step": 9
    },
    {
      "completion_length": 56.01041793823242,
      "epoch": 0.04987531172069826,
      "grad_norm": 41.213287353515625,
      "kl": 0.0005130767822265625,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 1.9375000596046448,
      "reward_std": 0.5250107049942017,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.6979166865348816,
      "rewards/value_reward": 0.2395833358168602,
      "step": 10
    },
    {
      "completion_length": 55.28125190734863,
      "epoch": 0.05486284289276808,
      "grad_norm": 2.5374794006347656,
      "kl": 0.0009784698486328125,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 1.78125,
      "reward_std": 0.5179632753133774,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.6354166865348816,
      "rewards/value_reward": 0.1458333358168602,
      "step": 11
    },
    {
      "completion_length": 51.43750190734863,
      "epoch": 0.059850374064837904,
      "grad_norm": 3.0466294288635254,
      "kl": 0.001834869384765625,
      "learning_rate": 1e-06,
      "loss": 0.0001,
      "reward": 1.6770833730697632,
      "reward_std": 0.5379315614700317,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.5625000298023224,
      "rewards/value_reward": 0.11458333395421505,
      "step": 12
    },
    {
      "completion_length": 47.427085876464844,
      "epoch": 0.06483790523690773,
      "grad_norm": 26.806480407714844,
      "kl": 0.0010318756103515625,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 1.7187500596046448,
      "reward_std": 0.4230456277728081,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.583333358168602,
      "rewards/value_reward": 0.1354166716337204,
      "step": 13
    },
    {
      "completion_length": 50.02083396911621,
      "epoch": 0.06982543640897755,
      "grad_norm": 1.8247559070587158,
      "kl": 0.0010204315185546875,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 1.9375001192092896,
      "reward_std": 0.4463787227869034,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.6979166865348816,
      "rewards/value_reward": 0.2395833432674408,
      "step": 14
    },
    {
      "completion_length": 52.68750190734863,
      "epoch": 0.07481296758104738,
      "grad_norm": 2.3722641468048096,
      "kl": 0.0010986328125,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 1.9062500596046448,
      "reward_std": 0.4055413603782654,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.6666666865348816,
      "rewards/value_reward": 0.2395833432674408,
      "step": 15
    },
    {
      "completion_length": 53.23958396911621,
      "epoch": 0.0798004987531172,
      "grad_norm": 3.7819693088531494,
      "kl": 0.001567840576171875,
      "learning_rate": 1e-06,
      "loss": 0.0001,
      "reward": 1.8645833730697632,
      "reward_std": 0.5996341705322266,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.6875000298023224,
      "rewards/value_reward": 0.1770833358168602,
      "step": 16
    },
    {
      "completion_length": 49.02083396911621,
      "epoch": 0.08478802992518704,
      "grad_norm": 2.7006280422210693,
      "kl": 0.00139617919921875,
      "learning_rate": 1e-06,
      "loss": 0.0001,
      "reward": 1.7187500596046448,
      "reward_std": 0.4375252425670624,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.6041666865348816,
      "rewards/value_reward": 0.1145833395421505,
      "step": 17
    },
    {
      "completion_length": 57.10416793823242,
      "epoch": 0.08977556109725686,
      "grad_norm": 10.572280883789062,
      "kl": 0.001865386962890625,
      "learning_rate": 1e-06,
      "loss": 0.0001,
      "reward": 1.7395833730697632,
      "reward_std": 0.5867283642292023,
      "rewards/format_reward": 0.9895833432674408,
      "rewards/type_reward": 0.5416666716337204,
      "rewards/value_reward": 0.2083333432674408,
      "step": 18
    },
    {
      "completion_length": 52.40625190734863,
      "epoch": 0.09476309226932668,
      "grad_norm": 3.5348286628723145,
      "kl": 0.00434112548828125,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.7187500596046448,
      "reward_std": 0.5008201450109482,
      "rewards/format_reward": 0.9895833432674408,
      "rewards/type_reward": 0.5833333730697632,
      "rewards/value_reward": 0.1458333395421505,
      "step": 19
    },
    {
      "completion_length": 54.54166793823242,
      "epoch": 0.09975062344139651,
      "grad_norm": 11.005522727966309,
      "kl": 0.00327301025390625,
      "learning_rate": 1e-06,
      "loss": 0.0001,
      "reward": 1.8645833730697632,
      "reward_std": 0.42722317576408386,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.6354166865348816,
      "rewards/value_reward": 0.2291666716337204,
      "step": 20
    },
    {
      "completion_length": 59.35416793823242,
      "epoch": 0.10473815461346633,
      "grad_norm": 3.138442039489746,
      "kl": 0.0045928955078125,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.8750000596046448,
      "reward_std": 0.4148429334163666,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.6666666865348816,
      "rewards/value_reward": 0.2083333358168602,
      "step": 21
    },
    {
      "completion_length": 57.17708396911621,
      "epoch": 0.10972568578553615,
      "grad_norm": 10.232250213623047,
      "kl": 0.001987457275390625,
      "learning_rate": 1e-06,
      "loss": 0.0001,
      "reward": 1.6145833730697632,
      "reward_std": 0.331511914730072,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.53125,
      "rewards/value_reward": 0.0833333358168602,
      "step": 22
    },
    {
      "completion_length": 55.29166793823242,
      "epoch": 0.11471321695760599,
      "grad_norm": 3.6366941928863525,
      "kl": 0.0028839111328125,
      "learning_rate": 1e-06,
      "loss": 0.0001,
      "reward": 1.7916666865348816,
      "reward_std": 0.48244550824165344,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.5937500298023224,
      "rewards/value_reward": 0.197916679084301,
      "step": 23
    },
    {
      "completion_length": 55.91666793823242,
      "epoch": 0.11970074812967581,
      "grad_norm": 4.003420829772949,
      "kl": 0.00351715087890625,
      "learning_rate": 1e-06,
      "loss": 0.0001,
      "reward": 2.0625001192092896,
      "reward_std": 0.5210920870304108,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.7812500298023224,
      "rewards/value_reward": 0.2812500074505806,
      "step": 24
    },
    {
      "completion_length": 52.16666793823242,
      "epoch": 0.12468827930174564,
      "grad_norm": 3.123845100402832,
      "kl": 0.00356292724609375,
      "learning_rate": 1e-06,
      "loss": 0.0001,
      "reward": 1.6875,
      "reward_std": 0.39690303802490234,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.5729166865348816,
      "rewards/value_reward": 0.1145833395421505,
      "step": 25
    },
    {
      "completion_length": 50.53125190734863,
      "epoch": 0.12967581047381546,
      "grad_norm": 2.6504132747650146,
      "kl": 0.0029296875,
      "learning_rate": 1e-06,
      "loss": 0.0001,
      "reward": 1.5833333730697632,
      "reward_std": 0.3844688832759857,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.5000000298023224,
      "rewards/value_reward": 0.0833333358168602,
      "step": 26
    },
    {
      "completion_length": 55.36458396911621,
      "epoch": 0.13466334164588528,
      "grad_norm": 3.807915210723877,
      "kl": 0.0046844482421875,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 2.020833373069763,
      "reward_std": 0.5475624799728394,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.6979166865348816,
      "rewards/value_reward": 0.3229166716337204,
      "step": 27
    },
    {
      "completion_length": 57.47916793823242,
      "epoch": 0.1396508728179551,
      "grad_norm": 5.694674491882324,
      "kl": 0.00274658203125,
      "learning_rate": 1e-06,
      "loss": 0.0001,
      "reward": 2.114583373069763,
      "reward_std": 0.4728766232728958,
      "rewards/format_reward": 0.9895833432674408,
      "rewards/type_reward": 0.7500000298023224,
      "rewards/value_reward": 0.375,
      "step": 28
    },
    {
      "completion_length": 55.41666793823242,
      "epoch": 0.14463840399002495,
      "grad_norm": 2.54262638092041,
      "kl": 0.00244140625,
      "learning_rate": 1e-06,
      "loss": 0.0001,
      "reward": 1.8125,
      "reward_std": 0.4684932827949524,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.6458333432674408,
      "rewards/value_reward": 0.1666666716337204,
      "step": 29
    },
    {
      "completion_length": 56.33333396911621,
      "epoch": 0.14962593516209477,
      "grad_norm": 3.026751756668091,
      "kl": 0.003482818603515625,
      "learning_rate": 1e-06,
      "loss": 0.0001,
      "reward": 2.1250001192092896,
      "reward_std": 0.5056642293930054,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.7083333432674408,
      "rewards/value_reward": 0.4166666865348816,
      "step": 30
    },
    {
      "completion_length": 48.06250190734863,
      "epoch": 0.1546134663341646,
      "grad_norm": 4.673459529876709,
      "kl": 0.00234222412109375,
      "learning_rate": 1e-06,
      "loss": 0.0001,
      "reward": 1.8437500596046448,
      "reward_std": 0.344024121761322,
      "rewards/format_reward": 0.9895833432674408,
      "rewards/type_reward": 0.7395833432674408,
      "rewards/value_reward": 0.1145833358168602,
      "step": 31
    },
    {
      "completion_length": 51.958335876464844,
      "epoch": 0.1596009975062344,
      "grad_norm": 3.077587366104126,
      "kl": 0.00388336181640625,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.7187500596046448,
      "reward_std": 0.46897804737091064,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.5416666716337204,
      "rewards/value_reward": 0.1770833432674408,
      "step": 32
    },
    {
      "completion_length": 53.21875190734863,
      "epoch": 0.16458852867830423,
      "grad_norm": 8.769026756286621,
      "kl": 0.00225830078125,
      "learning_rate": 1e-06,
      "loss": 0.0001,
      "reward": 1.8541667461395264,
      "reward_std": 0.3557152897119522,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.71875,
      "rewards/value_reward": 0.1354166716337204,
      "step": 33
    },
    {
      "completion_length": 52.69791793823242,
      "epoch": 0.16957605985037408,
      "grad_norm": 2.985428810119629,
      "kl": 0.00567626953125,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.9166666865348816,
      "reward_std": 0.39310361444950104,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.7083333730697632,
      "rewards/value_reward": 0.2083333432674408,
      "step": 34
    },
    {
      "completion_length": 51.50000190734863,
      "epoch": 0.1745635910224439,
      "grad_norm": 1.9351848363876343,
      "kl": 0.006256103515625,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.9166667461395264,
      "reward_std": 0.35025089979171753,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.6979166865348816,
      "rewards/value_reward": 0.21875,
      "step": 35
    },
    {
      "completion_length": 56.17708396911621,
      "epoch": 0.17955112219451372,
      "grad_norm": 5.072805404663086,
      "kl": 0.0033111572265625,
      "learning_rate": 1e-06,
      "loss": 0.0001,
      "reward": 2.114583373069763,
      "reward_std": 0.41180241107940674,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.7916666865348816,
      "rewards/value_reward": 0.3229166716337204,
      "step": 36
    },
    {
      "completion_length": 50.739585876464844,
      "epoch": 0.18453865336658354,
      "grad_norm": 12.998750686645508,
      "kl": 0.003429412841796875,
      "learning_rate": 1e-06,
      "loss": 0.0001,
      "reward": 2.0104167461395264,
      "reward_std": 0.2894075810909271,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.8229166865348816,
      "rewards/value_reward": 0.1875,
      "step": 37
    },
    {
      "completion_length": 52.97916793823242,
      "epoch": 0.18952618453865336,
      "grad_norm": 3.0100650787353516,
      "kl": 0.0068359375,
      "learning_rate": 1e-06,
      "loss": 0.0003,
      "reward": 1.7187500596046448,
      "reward_std": 0.4062570631504059,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.5520833432674408,
      "rewards/value_reward": 0.1666666716337204,
      "step": 38
    },
    {
      "completion_length": 50.12500190734863,
      "epoch": 0.19451371571072318,
      "grad_norm": 5.043798446655273,
      "kl": 0.008544921875,
      "learning_rate": 1e-06,
      "loss": 0.0003,
      "reward": 1.8437501192092896,
      "reward_std": 0.44689737260341644,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.614583358168602,
      "rewards/value_reward": 0.2291666679084301,
      "step": 39
    },
    {
      "completion_length": 52.46875190734863,
      "epoch": 0.19950124688279303,
      "grad_norm": 2.5372238159179688,
      "kl": 0.00316619873046875,
      "learning_rate": 1e-06,
      "loss": 0.0001,
      "reward": 2.1979167461395264,
      "reward_std": 0.31072379648685455,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.7395833432674408,
      "rewards/value_reward": 0.4583333432674408,
      "step": 40
    },
    {
      "completion_length": 57.85416793823242,
      "epoch": 0.20448877805486285,
      "grad_norm": 7.756105899810791,
      "kl": 0.022674560546875,
      "learning_rate": 1e-06,
      "loss": 0.0009,
      "reward": 1.7291667461395264,
      "reward_std": 0.5042311251163483,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.5000000149011612,
      "rewards/value_reward": 0.229166679084301,
      "step": 41
    },
    {
      "completion_length": 49.89583396911621,
      "epoch": 0.20947630922693267,
      "grad_norm": 7.1992363929748535,
      "kl": 0.00597381591796875,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.770833432674408,
      "reward_std": 0.36754731833934784,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.614583358168602,
      "rewards/value_reward": 0.15625,
      "step": 42
    },
    {
      "completion_length": 49.33333396911621,
      "epoch": 0.2144638403990025,
      "grad_norm": 3.374178886413574,
      "kl": 0.0060272216796875,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.75,
      "reward_std": 0.3439617305994034,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.5833333730697632,
      "rewards/value_reward": 0.1666666716337204,
      "step": 43
    },
    {
      "completion_length": 50.02083396911621,
      "epoch": 0.2194513715710723,
      "grad_norm": 2.0867555141448975,
      "kl": 0.015594482421875,
      "learning_rate": 1e-06,
      "loss": 0.0006,
      "reward": 2.0625,
      "reward_std": 0.3554982841014862,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.7187500298023224,
      "rewards/value_reward": 0.3437500149011612,
      "step": 44
    },
    {
      "completion_length": 51.29166793823242,
      "epoch": 0.22443890274314215,
      "grad_norm": 3.5470542907714844,
      "kl": 0.0106964111328125,
      "learning_rate": 1e-06,
      "loss": 0.0004,
      "reward": 1.7708333730697632,
      "reward_std": 0.34074608981609344,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.6145833432674408,
      "rewards/value_reward": 0.1562500037252903,
      "step": 45
    },
    {
      "completion_length": 50.625,
      "epoch": 0.22942643391521197,
      "grad_norm": 4.836516380310059,
      "kl": 0.00927734375,
      "learning_rate": 1e-06,
      "loss": 0.0004,
      "reward": 1.7604166865348816,
      "reward_std": 0.3411097154021263,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.6145833730697632,
      "rewards/value_reward": 0.1458333358168602,
      "step": 46
    },
    {
      "completion_length": 46.27083396911621,
      "epoch": 0.2344139650872818,
      "grad_norm": 2.7732608318328857,
      "kl": 0.013458251953125,
      "learning_rate": 1e-06,
      "loss": 0.0005,
      "reward": 1.7916667461395264,
      "reward_std": 0.36258386075496674,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.5625000298023224,
      "rewards/value_reward": 0.2291666716337204,
      "step": 47
    },
    {
      "completion_length": 46.39583396911621,
      "epoch": 0.23940149625935161,
      "grad_norm": 4.2999653816223145,
      "kl": 0.01378631591796875,
      "learning_rate": 1e-06,
      "loss": 0.0006,
      "reward": 1.7708333730697632,
      "reward_std": 0.31213822960853577,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.6979166865348816,
      "rewards/value_reward": 0.0729166716337204,
      "step": 48
    },
    {
      "completion_length": 52.26041793823242,
      "epoch": 0.24438902743142144,
      "grad_norm": 10.70241928100586,
      "kl": 0.0051727294921875,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 2.0104167461395264,
      "reward_std": 0.5115884244441986,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.7291666865348816,
      "rewards/value_reward": 0.28125,
      "step": 49
    },
    {
      "completion_length": 47.614585876464844,
      "epoch": 0.24937655860349128,
      "grad_norm": 5.545669078826904,
      "kl": 0.0100555419921875,
      "learning_rate": 1e-06,
      "loss": 0.0004,
      "reward": 1.8645833730697632,
      "reward_std": 0.3386664316058159,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.625,
      "rewards/value_reward": 0.2395833358168602,
      "step": 50
    },
    {
      "completion_length": 49.68750190734863,
      "epoch": 0.2543640897755611,
      "grad_norm": 3.6107425689697266,
      "kl": 0.0140380859375,
      "learning_rate": 1e-06,
      "loss": 0.0006,
      "reward": 2.2291667461395264,
      "reward_std": 0.3583437502384186,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.9375,
      "rewards/value_reward": 0.2916666716337204,
      "step": 51
    },
    {
      "completion_length": 53.51041793823242,
      "epoch": 0.2593516209476309,
      "grad_norm": 3.2154581546783447,
      "kl": 0.0047760009765625,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.6250000596046448,
      "reward_std": 0.27929287403821945,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.5,
      "rewards/value_reward": 0.12500000558793545,
      "step": 52
    },
    {
      "completion_length": 50.44791793823242,
      "epoch": 0.26433915211970077,
      "grad_norm": 1.3889214992523193,
      "kl": 0.004791259765625,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.96875,
      "reward_std": 0.21564548462629318,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.7187500298023224,
      "rewards/value_reward": 0.2500000074505806,
      "step": 53
    },
    {
      "completion_length": 49.26041793823242,
      "epoch": 0.26932668329177056,
      "grad_norm": 3.8615550994873047,
      "kl": 0.00653076171875,
      "learning_rate": 1e-06,
      "loss": 0.0003,
      "reward": 1.9791667461395264,
      "reward_std": 0.547309935092926,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.6041666865348816,
      "rewards/value_reward": 0.3750000149011612,
      "step": 54
    },
    {
      "completion_length": 53.65625190734863,
      "epoch": 0.2743142144638404,
      "grad_norm": 1.664948582649231,
      "kl": 0.00567626953125,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 2.2187501192092896,
      "reward_std": 0.3106548562645912,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.7604166865348816,
      "rewards/value_reward": 0.4583333432674408,
      "step": 55
    },
    {
      "completion_length": 52.58333396911621,
      "epoch": 0.2793017456359102,
      "grad_norm": 6.0326995849609375,
      "kl": 0.00933837890625,
      "learning_rate": 1e-06,
      "loss": 0.0004,
      "reward": 1.53125,
      "reward_std": 0.3061639815568924,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.46875,
      "rewards/value_reward": 0.0625,
      "step": 56
    },
    {
      "completion_length": 48.364585876464844,
      "epoch": 0.28428927680798005,
      "grad_norm": 6.530869007110596,
      "kl": 0.0138702392578125,
      "learning_rate": 1e-06,
      "loss": 0.0006,
      "reward": 1.6250000596046448,
      "reward_std": 0.33414995670318604,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.5312500149011612,
      "rewards/value_reward": 0.0937500037252903,
      "step": 57
    },
    {
      "completion_length": 49.51041793823242,
      "epoch": 0.2892768079800499,
      "grad_norm": 5.4448466300964355,
      "kl": 0.0064697265625,
      "learning_rate": 1e-06,
      "loss": 0.0003,
      "reward": 2.1562501192092896,
      "reward_std": 0.4056102931499481,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.7083333730697632,
      "rewards/value_reward": 0.4479166865348816,
      "step": 58
    },
    {
      "completion_length": 52.35416793823242,
      "epoch": 0.2942643391521197,
      "grad_norm": 6.877171039581299,
      "kl": 0.0279541015625,
      "learning_rate": 1e-06,
      "loss": 0.0011,
      "reward": 1.9583333730697632,
      "reward_std": 0.5338778495788574,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.6666666865348816,
      "rewards/value_reward": 0.291666679084301,
      "step": 59
    },
    {
      "completion_length": 49.78125,
      "epoch": 0.29925187032418954,
      "grad_norm": 1.9202797412872314,
      "kl": 0.0048370361328125,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 2.0000001192092896,
      "reward_std": 0.25321267172694206,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.8229166865348816,
      "rewards/value_reward": 0.1770833432674408,
      "step": 60
    },
    {
      "completion_length": 54.40625190734863,
      "epoch": 0.30423940149625933,
      "grad_norm": 5.348086833953857,
      "kl": 0.0098114013671875,
      "learning_rate": 1e-06,
      "loss": 0.0004,
      "reward": 1.4270833730697632,
      "reward_std": 0.2687583640217781,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.4062500149011612,
      "rewards/value_reward": 0.02083333395421505,
      "step": 61
    },
    {
      "completion_length": 47.12500190734863,
      "epoch": 0.3092269326683292,
      "grad_norm": 2.447073221206665,
      "kl": 0.0063934326171875,
      "learning_rate": 1e-06,
      "loss": 0.0003,
      "reward": 1.7291666865348816,
      "reward_std": 0.26041998714208603,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.5208333432674408,
      "rewards/value_reward": 0.2083333358168602,
      "step": 62
    },
    {
      "completion_length": 45.76041793823242,
      "epoch": 0.314214463840399,
      "grad_norm": 4.828869342803955,
      "kl": 0.0174102783203125,
      "learning_rate": 1e-06,
      "loss": 0.0007,
      "reward": 2.114583373069763,
      "reward_std": 0.45097219944000244,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.7291666865348816,
      "rewards/value_reward": 0.3854166716337204,
      "step": 63
    },
    {
      "completion_length": 47.98958396911621,
      "epoch": 0.3192019950124688,
      "grad_norm": 4.6910271644592285,
      "kl": 0.04632568359375,
      "learning_rate": 1e-06,
      "loss": 0.0018,
      "reward": 1.9479167461395264,
      "reward_std": 0.3890128433704376,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.6666666865348816,
      "rewards/value_reward": 0.2812500074505806,
      "step": 64
    },
    {
      "completion_length": 53.79166793823242,
      "epoch": 0.32418952618453867,
      "grad_norm": 3.737656354904175,
      "kl": 0.0073089599609375,
      "learning_rate": 1e-06,
      "loss": 0.0003,
      "reward": 1.6979167461395264,
      "reward_std": 0.3460918813943863,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.4895833432674408,
      "rewards/value_reward": 0.2083333358168602,
      "step": 65
    },
    {
      "completion_length": 52.50000190734863,
      "epoch": 0.32917705735660846,
      "grad_norm": 2.92228627204895,
      "kl": 0.0075836181640625,
      "learning_rate": 1e-06,
      "loss": 0.0003,
      "reward": 2.1145834922790527,
      "reward_std": 0.40096497535705566,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.8125,
      "rewards/value_reward": 0.3020833432674408,
      "step": 66
    },
    {
      "completion_length": 52.91666793823242,
      "epoch": 0.3341645885286783,
      "grad_norm": 7.9724273681640625,
      "kl": 0.0267333984375,
      "learning_rate": 1e-06,
      "loss": 0.0011,
      "reward": 2.114583373069763,
      "reward_std": 0.6110064685344696,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.7812500298023224,
      "rewards/value_reward": 0.3333333432674408,
      "step": 67
    },
    {
      "completion_length": 50.520835876464844,
      "epoch": 0.33915211970074816,
      "grad_norm": 4.161324977874756,
      "kl": 0.01348876953125,
      "learning_rate": 1e-06,
      "loss": 0.0005,
      "reward": 1.7812500596046448,
      "reward_std": 0.34851498901844025,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.6562500298023224,
      "rewards/value_reward": 0.125,
      "step": 68
    },
    {
      "completion_length": 44.19791793823242,
      "epoch": 0.34413965087281795,
      "grad_norm": 3.0720622539520264,
      "kl": 0.0056304931640625,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.927083432674408,
      "reward_std": 0.2689402103424072,
      "rewards/format_reward": 0.9895833432674408,
      "rewards/type_reward": 0.7083333432674408,
      "rewards/value_reward": 0.2291666679084301,
      "step": 69
    },
    {
      "completion_length": 50.63541793823242,
      "epoch": 0.3491271820448878,
      "grad_norm": 4.674132823944092,
      "kl": 0.0147705078125,
      "learning_rate": 1e-06,
      "loss": 0.0006,
      "reward": 2.3020834922790527,
      "reward_std": 0.5566646903753281,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.8020833432674408,
      "rewards/value_reward": 0.5000000298023224,
      "step": 70
    },
    {
      "completion_length": 41.88541793823242,
      "epoch": 0.3541147132169576,
      "grad_norm": 2.771658182144165,
      "kl": 0.059356689453125,
      "learning_rate": 1e-06,
      "loss": 0.0024,
      "reward": 1.78125,
      "reward_std": 0.20404693484306335,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.6458333730697632,
      "rewards/value_reward": 0.1354166716337204,
      "step": 71
    },
    {
      "completion_length": 52.1875,
      "epoch": 0.35910224438902744,
      "grad_norm": 2.4222514629364014,
      "kl": 0.005126953125,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.7395833730697632,
      "reward_std": 0.1746465563774109,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.6041666865348816,
      "rewards/value_reward": 0.1354166716337204,
      "step": 72
    },
    {
      "completion_length": 50.60416793823242,
      "epoch": 0.3640897755610973,
      "grad_norm": 3.850780487060547,
      "kl": 0.009765625,
      "learning_rate": 1e-06,
      "loss": 0.0004,
      "reward": 2.145833432674408,
      "reward_std": 0.38716475665569305,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.6979166865348816,
      "rewards/value_reward": 0.4479166716337204,
      "step": 73
    },
    {
      "completion_length": 52.55208396911621,
      "epoch": 0.3690773067331671,
      "grad_norm": 4.4836297035217285,
      "kl": 0.020843505859375,
      "learning_rate": 1e-06,
      "loss": 0.0008,
      "reward": 2.0312501192092896,
      "reward_std": 0.3606877475976944,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.7083333432674408,
      "rewards/value_reward": 0.3229166716337204,
      "step": 74
    },
    {
      "completion_length": 49.59375190734863,
      "epoch": 0.3740648379052369,
      "grad_norm": 5.496286392211914,
      "kl": 0.005767822265625,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.9166666865348816,
      "reward_std": 0.3684488981962204,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.7291666865348816,
      "rewards/value_reward": 0.1875000074505806,
      "step": 75
    },
    {
      "completion_length": 46.41666793823242,
      "epoch": 0.3790523690773067,
      "grad_norm": 9.9849271774292,
      "kl": 0.014251708984375,
      "learning_rate": 1e-06,
      "loss": 0.0006,
      "reward": 2.03125,
      "reward_std": 0.3265463262796402,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.6041666865348816,
      "rewards/value_reward": 0.4270833432674408,
      "step": 76
    },
    {
      "completion_length": 50.34375190734863,
      "epoch": 0.38403990024937656,
      "grad_norm": 4.3121018409729,
      "kl": 0.009674072265625,
      "learning_rate": 1e-06,
      "loss": 0.0004,
      "reward": 2.0000001192092896,
      "reward_std": 0.2965727299451828,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.7187500298023224,
      "rewards/value_reward": 0.2812500111758709,
      "step": 77
    },
    {
      "completion_length": 50.01041793823242,
      "epoch": 0.38902743142144636,
      "grad_norm": 1.8322099447250366,
      "kl": 0.0094757080078125,
      "learning_rate": 1e-06,
      "loss": 0.0004,
      "reward": 2.1875,
      "reward_std": 0.22901885211467743,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.8541666865348816,
      "rewards/value_reward": 0.3333333432674408,
      "step": 78
    },
    {
      "completion_length": 46.48958396911621,
      "epoch": 0.3940149625935162,
      "grad_norm": 4.674933910369873,
      "kl": 0.0077667236328125,
      "learning_rate": 1e-06,
      "loss": 0.0003,
      "reward": 2.2291667461395264,
      "reward_std": 0.41207581758499146,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.8333333432674408,
      "rewards/value_reward": 0.395833358168602,
      "step": 79
    },
    {
      "completion_length": 48.14583396911621,
      "epoch": 0.39900249376558605,
      "grad_norm": 7.052217483520508,
      "kl": 0.016845703125,
      "learning_rate": 1e-06,
      "loss": 0.0007,
      "reward": 2.1666667461395264,
      "reward_std": 0.3924287408590317,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.7812500298023224,
      "rewards/value_reward": 0.3854166865348816,
      "step": 80
    },
    {
      "completion_length": 54.76041793823242,
      "epoch": 0.40399002493765584,
      "grad_norm": 6.566869735717773,
      "kl": 0.022216796875,
      "learning_rate": 1e-06,
      "loss": 0.0009,
      "reward": 2.4687501192092896,
      "reward_std": 0.48340874910354614,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.8854166865348816,
      "rewards/value_reward": 0.5833333730697632,
      "step": 81
    },
    {
      "completion_length": 46.19791793823242,
      "epoch": 0.4089775561097257,
      "grad_norm": 2.225270986557007,
      "kl": 0.01568603515625,
      "learning_rate": 1e-06,
      "loss": 0.0006,
      "reward": 1.7812500596046448,
      "reward_std": 0.31752340495586395,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.4791666865348816,
      "rewards/value_reward": 0.3020833432674408,
      "step": 82
    },
    {
      "completion_length": 45.9375,
      "epoch": 0.4139650872817955,
      "grad_norm": 2.7589635848999023,
      "kl": 0.02484130859375,
      "learning_rate": 1e-06,
      "loss": 0.001,
      "reward": 2.0625001192092896,
      "reward_std": 0.15642696991562843,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.8333333432674408,
      "rewards/value_reward": 0.229166679084301,
      "step": 83
    },
    {
      "completion_length": 47.48958396911621,
      "epoch": 0.41895261845386533,
      "grad_norm": 7.477308750152588,
      "kl": 0.014617919921875,
      "learning_rate": 1e-06,
      "loss": 0.0006,
      "reward": 2.0833334922790527,
      "reward_std": 0.2517358362674713,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.7604166865348816,
      "rewards/value_reward": 0.3229166716337204,
      "step": 84
    },
    {
      "completion_length": 46.125,
      "epoch": 0.4239401496259352,
      "grad_norm": 5.973231792449951,
      "kl": 0.00994873046875,
      "learning_rate": 1e-06,
      "loss": 0.0004,
      "reward": 2.1354167461395264,
      "reward_std": 0.23450180888175964,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.7604166865348816,
      "rewards/value_reward": 0.3750000223517418,
      "step": 85
    },
    {
      "completion_length": 50.36458396911621,
      "epoch": 0.428927680798005,
      "grad_norm": 2.558129072189331,
      "kl": 0.00982666015625,
      "learning_rate": 1e-06,
      "loss": 0.0004,
      "reward": 2.270833373069763,
      "reward_std": 0.2335097175091505,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.8125000298023224,
      "rewards/value_reward": 0.4583333358168602,
      "step": 86
    },
    {
      "completion_length": 52.71875190734863,
      "epoch": 0.4339152119700748,
      "grad_norm": 5.039067268371582,
      "kl": 0.01031494140625,
      "learning_rate": 1e-06,
      "loss": 0.0004,
      "reward": 1.958333432674408,
      "reward_std": 0.30705152451992035,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.6770833432674408,
      "rewards/value_reward": 0.2812500149011612,
      "step": 87
    },
    {
      "completion_length": 46.66666793823242,
      "epoch": 0.4389027431421446,
      "grad_norm": 3.2081820964813232,
      "kl": 0.0118255615234375,
      "learning_rate": 1e-06,
      "loss": 0.0005,
      "reward": 2.1979167461395264,
      "reward_std": 0.29271166026592255,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.8020833730697632,
      "rewards/value_reward": 0.3958333432674408,
      "step": 88
    },
    {
      "completion_length": 43.85416793823242,
      "epoch": 0.44389027431421446,
      "grad_norm": 3.229139566421509,
      "kl": 0.017974853515625,
      "learning_rate": 1e-06,
      "loss": 0.0007,
      "reward": 1.9687501192092896,
      "reward_std": 0.34249694645404816,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.8125,
      "rewards/value_reward": 0.15625000558793545,
      "step": 89
    },
    {
      "completion_length": 41.66666793823242,
      "epoch": 0.4488778054862843,
      "grad_norm": 12.049365997314453,
      "kl": 0.012603759765625,
      "learning_rate": 1e-06,
      "loss": 0.0005,
      "reward": 2.0937500596046448,
      "reward_std": 0.29995445907115936,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.7708333432674408,
      "rewards/value_reward": 0.3229166716337204,
      "step": 90
    },
    {
      "completion_length": 44.58333396911621,
      "epoch": 0.4538653366583541,
      "grad_norm": 7.588222980499268,
      "kl": 0.038848876953125,
      "learning_rate": 1e-06,
      "loss": 0.0015,
      "reward": 1.8750001192092896,
      "reward_std": 0.2700177952647209,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.6562500298023224,
      "rewards/value_reward": 0.21875,
      "step": 91
    },
    {
      "completion_length": 50.17708396911621,
      "epoch": 0.45885286783042395,
      "grad_norm": 1.2638224363327026,
      "kl": 0.0105743408203125,
      "learning_rate": 1e-06,
      "loss": 0.0004,
      "reward": 1.895833432674408,
      "reward_std": 0.15039033815264702,
      "rewards/format_reward": 0.9895833432674408,
      "rewards/type_reward": 0.75,
      "rewards/value_reward": 0.15625000558793545,
      "step": 92
    },
    {
      "completion_length": 51.60416793823242,
      "epoch": 0.46384039900249374,
      "grad_norm": 4.8879313468933105,
      "kl": 0.012725830078125,
      "learning_rate": 1e-06,
      "loss": 0.0005,
      "reward": 1.8958333730697632,
      "reward_std": 0.3485773950815201,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.6562500298023224,
      "rewards/value_reward": 0.2395833358168602,
      "step": 93
    },
    {
      "completion_length": 44.50000190734863,
      "epoch": 0.4688279301745636,
      "grad_norm": 1.3869191408157349,
      "kl": 0.0136566162109375,
      "learning_rate": 1e-06,
      "loss": 0.0005,
      "reward": 2.21875,
      "reward_std": 0.23043328523635864,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.8958333730697632,
      "rewards/value_reward": 0.322916679084301,
      "step": 94
    },
    {
      "completion_length": 46.63541793823242,
      "epoch": 0.47381546134663344,
      "grad_norm": 2.261981725692749,
      "kl": 0.01190185546875,
      "learning_rate": 1e-06,
      "loss": 0.0005,
      "reward": 2.2500001192092896,
      "reward_std": 0.19500280916690826,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.8333333730697632,
      "rewards/value_reward": 0.4166666716337204,
      "step": 95
    },
    {
      "completion_length": 42.39583396911621,
      "epoch": 0.47880299251870323,
      "grad_norm": 6.752196311950684,
      "kl": 0.01824951171875,
      "learning_rate": 1e-06,
      "loss": 0.0007,
      "reward": 1.7291667461395264,
      "reward_std": 0.17677670530974865,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.5833333432674408,
      "rewards/value_reward": 0.1458333432674408,
      "step": 96
    },
    {
      "completion_length": 44.520835876464844,
      "epoch": 0.4837905236907731,
      "grad_norm": 3.292236566543579,
      "kl": 0.011383056640625,
      "learning_rate": 1e-06,
      "loss": 0.0005,
      "reward": 2.1979167461395264,
      "reward_std": 0.30453013628721237,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.8437500298023224,
      "rewards/value_reward": 0.3541666716337204,
      "step": 97
    },
    {
      "completion_length": 44.177085876464844,
      "epoch": 0.48877805486284287,
      "grad_norm": 1.9719746112823486,
      "kl": 0.017486572265625,
      "learning_rate": 1e-06,
      "loss": 0.0007,
      "reward": 2.0729166865348816,
      "reward_std": 0.2895440012216568,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.6979166865348816,
      "rewards/value_reward": 0.3750000149011612,
      "step": 98
    },
    {
      "completion_length": 44.875,
      "epoch": 0.4937655860349127,
      "grad_norm": 2.0317137241363525,
      "kl": 0.015380859375,
      "learning_rate": 1e-06,
      "loss": 0.0006,
      "reward": 2.1458334922790527,
      "reward_std": 0.24789174646139145,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.7812500298023224,
      "rewards/value_reward": 0.3645833432674408,
      "step": 99
    },
    {
      "completion_length": 45.78125190734863,
      "epoch": 0.49875311720698257,
      "grad_norm": 2.7145822048187256,
      "kl": 0.02294921875,
      "learning_rate": 1e-06,
      "loss": 0.0009,
      "reward": 2.239583373069763,
      "reward_std": 0.2199057899415493,
      "rewards/format_reward": 1.0,
      "rewards/type_reward": 0.8958333730697632,
      "rewards/value_reward": 0.3437500074505806,
      "step": 100
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 200,
  "num_input_tokens_seen": 0,
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}