{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9903536977491961,
  "eval_steps": 500,
  "global_step": 77,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "clip_ratio": 0.0,
      "completion_length": 1084.34375,
      "epoch": 0.012861736334405145,
      "grad_norm": 0.41294848918914795,
      "kl": 0.0,
      "learning_rate": 1.25e-07,
      "loss": 0.053,
      "reward": 1.54296875,
      "reward_std": 0.11134403571486473,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.54296875,
      "step": 1
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 912.421875,
      "epoch": 0.02572347266881029,
      "grad_norm": 0.5866098403930664,
      "kl": 0.0,
      "learning_rate": 2.5e-07,
      "loss": 0.1483,
      "reward": 1.5625,
      "reward_std": 0.15145022049546242,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.5625,
      "step": 2
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1136.859375,
      "epoch": 0.03858520900321544,
      "grad_norm": 0.5072430372238159,
      "kl": 0.00023818016052246094,
      "learning_rate": 3.75e-07,
      "loss": 0.117,
      "reward": 1.57421875,
      "reward_std": 0.1861907932907343,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.57421875,
      "step": 3
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1155.015625,
      "epoch": 0.05144694533762058,
      "grad_norm": 0.4072016179561615,
      "kl": 0.00027489662170410156,
      "learning_rate": 5e-07,
      "loss": 0.0995,
      "reward": 1.5390625,
      "reward_std": 0.10738485306501389,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.5390625,
      "step": 4
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 966.703125,
      "epoch": 0.06430868167202572,
      "grad_norm": 0.5740306377410889,
      "kl": 0.0003867149353027344,
      "learning_rate": 6.249999999999999e-07,
      "loss": 0.1656,
      "reward": 1.57421875,
      "reward_std": 0.11065823957324028,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.57421875,
      "step": 5
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1206.765625,
      "epoch": 0.07717041800643087,
      "grad_norm": 0.4439733028411865,
      "kl": 0.00043845176696777344,
      "learning_rate": 7.5e-07,
      "loss": 0.0946,
      "reward": 1.515625,
      "reward_std": 0.08086910098791122,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.515625,
      "step": 6
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1010.46875,
      "epoch": 0.09003215434083602,
      "grad_norm": 0.2982307970523834,
      "kl": 0.00031566619873046875,
      "learning_rate": 8.75e-07,
      "loss": 0.0943,
      "reward": 1.53125,
      "reward_std": 0.09774631634354591,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.53125,
      "step": 7
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1046.9375,
      "epoch": 0.10289389067524116,
      "grad_norm": 0.4670819938182831,
      "kl": 0.00023698806762695312,
      "learning_rate": 1e-06,
      "loss": 0.1583,
      "reward": 1.5390625,
      "reward_std": 0.13101092725992203,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.5390625,
      "step": 8
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 966.28125,
      "epoch": 0.1157556270096463,
      "grad_norm": 0.40219369530677795,
      "kl": 0.0002582073211669922,
      "learning_rate": 9.995336531410273e-07,
      "loss": 0.1739,
      "reward": 1.546875,
      "reward_std": 0.1388406604528427,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.546875,
      "step": 9
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1078.390625,
      "epoch": 0.12861736334405144,
      "grad_norm": 0.291391521692276,
      "kl": 0.00041747093200683594,
      "learning_rate": 9.98135579139189e-07,
      "loss": -0.0025,
      "reward": 1.53125,
      "reward_std": 0.04564354941248894,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.53125,
      "step": 10
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1055.640625,
      "epoch": 0.1414790996784566,
      "grad_norm": 0.34915173053741455,
      "kl": 0.00037789344787597656,
      "learning_rate": 9.958086757163488e-07,
      "loss": 0.0703,
      "reward": 1.56640625,
      "reward_std": 0.10219397768378258,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.56640625,
      "step": 11
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1095.640625,
      "epoch": 0.15434083601286175,
      "grad_norm": 0.550199031829834,
      "kl": 0.0002205371856689453,
      "learning_rate": 9.92557765735184e-07,
      "loss": 0.2093,
      "reward": 1.58203125,
      "reward_std": 0.19091684743762016,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.58203125,
      "step": 12
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1095.921875,
      "epoch": 0.16720257234726688,
      "grad_norm": 0.4369826316833496,
      "kl": 0.00036525726318359375,
      "learning_rate": 9.883895872030657e-07,
      "loss": 0.1704,
      "reward": 1.52734375,
      "reward_std": 0.09039457142353058,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.52734375,
      "step": 13
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 837.65625,
      "epoch": 0.18006430868167203,
      "grad_norm": 0.25077515840530396,
      "kl": 0.0001766681671142578,
      "learning_rate": 9.833127793065097e-07,
      "loss": 0.0155,
      "reward": 1.50390625,
      "reward_std": 0.015625,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.50390625,
      "step": 14
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1183.109375,
      "epoch": 0.19292604501607716,
      "grad_norm": 0.42581164836883545,
      "kl": 0.00038433074951171875,
      "learning_rate": 9.773378645051436e-07,
      "loss": 0.1642,
      "reward": 1.5,
      "reward_std": 0.06994619965553284,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.5,
      "step": 15
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1042.71875,
      "epoch": 0.2057877813504823,
      "grad_norm": 0.371883749961853,
      "kl": 0.00041294097900390625,
      "learning_rate": 9.704772267223019e-07,
      "loss": 0.0703,
      "reward": 1.54296875,
      "reward_std": 0.08989579975605011,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.54296875,
      "step": 16
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 978.828125,
      "epoch": 0.21864951768488747,
      "grad_norm": 0.4985962212085724,
      "kl": 0.00021839141845703125,
      "learning_rate": 9.627450856774539e-07,
      "loss": 0.079,
      "reward": 1.5703125,
      "reward_std": 0.11574668437242508,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.5703125,
      "step": 17
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 850.4375,
      "epoch": 0.2315112540192926,
      "grad_norm": 0.49231991171836853,
      "kl": 0.0002932548522949219,
      "learning_rate": 9.541574674136632e-07,
      "loss": 0.0227,
      "reward": 1.53125,
      "reward_std": 0.0816391110420227,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.53125,
      "step": 18
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1099.109375,
      "epoch": 0.24437299035369775,
      "grad_norm": 0.5843875408172607,
      "kl": 0.0002193450927734375,
      "learning_rate": 9.447321710811674e-07,
      "loss": 0.1417,
      "reward": 1.5390625,
      "reward_std": 0.1500675454735756,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.5390625,
      "step": 19
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1058.03125,
      "epoch": 0.2572347266881029,
      "grad_norm": 0.6623404026031494,
      "kl": 0.0004153251647949219,
      "learning_rate": 9.344887320459198e-07,
      "loss": 0.1164,
      "reward": 1.55078125,
      "reward_std": 0.10405365750193596,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.55078125,
      "step": 20
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 820.734375,
      "epoch": 0.27009646302250806,
      "grad_norm": 0.6680777072906494,
      "kl": 0.0003333091735839844,
      "learning_rate": 9.234483813995613e-07,
      "loss": 0.2424,
      "reward": 1.55078125,
      "reward_std": 0.174929304048419,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.55078125,
      "step": 21
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1022.640625,
      "epoch": 0.2829581993569132,
      "grad_norm": 0.6333020329475403,
      "kl": 0.0004470348358154297,
      "learning_rate": 9.116340019547401e-07,
      "loss": 0.0632,
      "reward": 1.51171875,
      "reward_std": 0.08957062661647797,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.51171875,
      "step": 22
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1067.015625,
      "epoch": 0.2958199356913183,
      "grad_norm": 0.5886364579200745,
      "kl": 0.0003254413604736328,
      "learning_rate": 8.990700808169889e-07,
      "loss": 0.1254,
      "reward": 1.5703125,
      "reward_std": 0.20152484998106956,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.5703125,
      "step": 23
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1248.078125,
      "epoch": 0.3086816720257235,
      "grad_norm": 0.6907972693443298,
      "kl": 0.0006000995635986328,
      "learning_rate": 8.857826586314586e-07,
      "loss": 0.0326,
      "reward": 1.54296875,
      "reward_std": 0.13257849216461182,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.54296875,
      "step": 24
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1049.546875,
      "epoch": 0.3215434083601286,
      "grad_norm": 0.45920121669769287,
      "kl": 0.0003612041473388672,
      "learning_rate": 8.717992756097047e-07,
      "loss": 0.0834,
      "reward": 1.5546875,
      "reward_std": 0.1335771605372429,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.5546875,
      "step": 25
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 912.46875,
      "epoch": 0.33440514469453375,
      "grad_norm": 0.336818665266037,
      "kl": 0.00023627281188964844,
      "learning_rate": 8.571489144483944e-07,
      "loss": 0.04,
      "reward": 1.51953125,
      "reward_std": 0.06524410098791122,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.51953125,
      "step": 26
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1066.484375,
      "epoch": 0.34726688102893893,
      "grad_norm": 0.39180704951286316,
      "kl": 0.00023651123046875,
      "learning_rate": 8.418619402582402e-07,
      "loss": 0.1659,
      "reward": 1.546875,
      "reward_std": 0.12253788113594055,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.546875,
      "step": 27
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1128.390625,
      "epoch": 0.36012861736334406,
      "grad_norm": 0.7383729815483093,
      "kl": 0.0007467269897460938,
      "learning_rate": 8.259700376276723e-07,
      "loss": 0.15,
      "reward": 1.578125,
      "reward_std": 0.14458869770169258,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.578125,
      "step": 28
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1031.703125,
      "epoch": 0.3729903536977492,
      "grad_norm": 0.730670154094696,
      "kl": 0.0005695819854736328,
      "learning_rate": 8.095061449516902e-07,
      "loss": -0.0424,
      "reward": 1.55859375,
      "reward_std": 0.11572204157710075,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.55859375,
      "step": 29
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1088.359375,
      "epoch": 0.3858520900321543,
      "grad_norm": 0.4401358366012573,
      "kl": 0.0004668235778808594,
      "learning_rate": 7.92504386162009e-07,
      "loss": 0.0101,
      "reward": 1.58984375,
      "reward_std": 0.10618764162063599,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.58984375,
      "step": 30
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 970.046875,
      "epoch": 0.3987138263665595,
      "grad_norm": 0.57388836145401,
      "kl": 0.00038433074951171875,
      "learning_rate": 7.75e-07,
      "loss": 0.0789,
      "reward": 1.64453125,
      "reward_std": 0.18705939501523972,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.64453125,
      "step": 31
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 852.6875,
      "epoch": 0.4115755627009646,
      "grad_norm": 0.5620797872543335,
      "kl": 0.0002961158752441406,
      "learning_rate": 7.570292669790184e-07,
      "loss": 0.0866,
      "reward": 1.6171875,
      "reward_std": 0.21375522762537003,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.6171875,
      "step": 32
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1189.84375,
      "epoch": 0.42443729903536975,
      "grad_norm": 0.4780595302581787,
      "kl": 0.00036525726318359375,
      "learning_rate": 7.38629434187499e-07,
      "loss": -0.0424,
      "reward": 1.5546875,
      "reward_std": 0.10984911397099495,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.5546875,
      "step": 33
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1082.984375,
      "epoch": 0.43729903536977494,
      "grad_norm": 0.6845836043357849,
      "kl": 0.0010368824005126953,
      "learning_rate": 7.198386380886764e-07,
      "loss": 0.0489,
      "reward": 1.57421875,
      "reward_std": 0.15323764458298683,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.57421875,
      "step": 34
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1034.34375,
      "epoch": 0.45016077170418006,
      "grad_norm": 0.46563801169395447,
      "kl": 0.0003986358642578125,
      "learning_rate": 7.006958254769437e-07,
      "loss": 0.166,
      "reward": 1.57421875,
      "reward_std": 0.13553934916853905,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.57421875,
      "step": 35
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 918.84375,
      "epoch": 0.4630225080385852,
      "grad_norm": 0.5318493247032166,
      "kl": 0.0004329681396484375,
      "learning_rate": 6.812406727546712e-07,
      "loss": 0.0854,
      "reward": 1.5625,
      "reward_std": 0.1329439952969551,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.5625,
      "step": 36
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 984.84375,
      "epoch": 0.4758842443729904,
      "grad_norm": 0.5245525240898132,
      "kl": 0.00041675567626953125,
      "learning_rate": 6.615135036968049e-07,
      "loss": 0.1983,
      "reward": 1.640625,
      "reward_std": 0.18561063334345818,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.015625,
      "rewards/tag_count_reward": 0.625,
      "step": 37
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 948.4375,
      "epoch": 0.4887459807073955,
      "grad_norm": 0.617463231086731,
      "kl": 0.0012621879577636719,
      "learning_rate": 6.415552058736853e-07,
      "loss": 0.1642,
      "reward": 1.65234375,
      "reward_std": 0.205689437687397,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.65234375,
      "step": 38
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 824.5,
      "epoch": 0.5016077170418006,
      "grad_norm": 0.6972614526748657,
      "kl": 0.000629425048828125,
      "learning_rate": 6.21407145905313e-07,
      "loss": 0.1105,
      "reward": 1.6484375,
      "reward_std": 0.2235238291323185,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.6484375,
      "step": 39
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1114.5,
      "epoch": 0.5144694533762058,
      "grad_norm": 0.5985956788063049,
      "kl": 0.0006375312805175781,
      "learning_rate": 6.011110837227137e-07,
      "loss": 0.1203,
      "reward": 1.6015625,
      "reward_std": 0.17044181004166603,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.6015625,
      "step": 40
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 884.921875,
      "epoch": 0.5273311897106109,
      "grad_norm": 0.5041890740394592,
      "kl": 0.000698089599609375,
      "learning_rate": 5.80709086014102e-07,
      "loss": 0.1753,
      "reward": 1.609375,
      "reward_std": 0.19224052503705025,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.609375,
      "step": 41
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1000.265625,
      "epoch": 0.5401929260450161,
      "grad_norm": 0.5194684267044067,
      "kl": 0.000476837158203125,
      "learning_rate": 5.602434390352476e-07,
      "loss": 0.1213,
      "reward": 1.6015625,
      "reward_std": 0.19315097481012344,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.6015625,
      "step": 42
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 973.03125,
      "epoch": 0.5530546623794212,
      "grad_norm": 0.6358352303504944,
      "kl": 0.0004420280456542969,
      "learning_rate": 5.397565609647524e-07,
      "loss": 0.1533,
      "reward": 1.671875,
      "reward_std": 0.20554664731025696,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.671875,
      "step": 43
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1015.921875,
      "epoch": 0.5659163987138264,
      "grad_norm": 0.5505901575088501,
      "kl": 0.0008034706115722656,
      "learning_rate": 5.192909139858981e-07,
      "loss": 0.1629,
      "reward": 1.59375,
      "reward_std": 0.16546358913183212,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.59375,
      "step": 44
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 726.28125,
      "epoch": 0.5787781350482315,
      "grad_norm": 0.6235621571540833,
      "kl": 0.0012774467468261719,
      "learning_rate": 4.988889162772862e-07,
      "loss": 0.2198,
      "reward": 1.640625,
      "reward_std": 0.19715734012424946,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.640625,
      "step": 45
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1159.125,
      "epoch": 0.5916398713826366,
      "grad_norm": 0.5912997722625732,
      "kl": 0.0010008811950683594,
      "learning_rate": 4.785928540946868e-07,
      "loss": 0.179,
      "reward": 1.56640625,
      "reward_std": 0.12528811395168304,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.56640625,
      "step": 46
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1131.15625,
      "epoch": 0.6045016077170418,
      "grad_norm": 0.521943986415863,
      "kl": 0.0008096694946289062,
      "learning_rate": 4.584447941263149e-07,
      "loss": 0.0605,
      "reward": 1.58984375,
      "reward_std": 0.12631270289421082,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.58984375,
      "step": 47
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1118.828125,
      "epoch": 0.617363344051447,
      "grad_norm": 0.4915071725845337,
      "kl": 0.0010766983032226562,
      "learning_rate": 4.384864963031951e-07,
      "loss": 0.2512,
      "reward": 1.6484375,
      "reward_std": 0.2036767303943634,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.6484375,
      "step": 48
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1062.84375,
      "epoch": 0.6302250803858521,
      "grad_norm": 0.529114305973053,
      "kl": 0.0007066726684570312,
      "learning_rate": 4.187593272453288e-07,
      "loss": 0.1496,
      "reward": 1.60546875,
      "reward_std": 0.15380773693323135,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.60546875,
      "step": 49
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 694.140625,
      "epoch": 0.6430868167202572,
      "grad_norm": 0.6105093359947205,
      "kl": 0.0016994476318359375,
      "learning_rate": 3.993041745230562e-07,
      "loss": 0.3051,
      "reward": 1.203125,
      "reward_std": 0.2077712006866932,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.703125,
      "step": 50
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 982.140625,
      "epoch": 0.6559485530546624,
      "grad_norm": 0.540160596370697,
      "kl": 0.0007996559143066406,
      "learning_rate": 3.8016136191132354e-07,
      "loss": 0.0934,
      "reward": 1.609375,
      "reward_std": 0.1588139720261097,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.609375,
      "step": 51
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 937.734375,
      "epoch": 0.6688102893890675,
      "grad_norm": 0.6522411108016968,
      "kl": 0.0015106201171875,
      "learning_rate": 3.6137056581250137e-07,
      "loss": 0.1704,
      "reward": 1.6171875,
      "reward_std": 0.1773756481707096,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.6171875,
      "step": 52
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 831.9375,
      "epoch": 0.6816720257234726,
      "grad_norm": 0.6507072448730469,
      "kl": 0.0011072158813476562,
      "learning_rate": 3.4297073302098155e-07,
      "loss": 0.1508,
      "reward": 1.65625,
      "reward_std": 0.1978173814713955,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.65625,
      "step": 53
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1121.0625,
      "epoch": 0.6945337620578779,
      "grad_norm": 0.5701295137405396,
      "kl": 0.000888824462890625,
      "learning_rate": 3.250000000000001e-07,
      "loss": 0.204,
      "reward": 1.65234375,
      "reward_std": 0.20623698085546494,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.65234375,
      "step": 54
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 880.375,
      "epoch": 0.707395498392283,
      "grad_norm": 0.7120354175567627,
      "kl": 0.0023937225341796875,
      "learning_rate": 3.0749561383799107e-07,
      "loss": 0.1468,
      "reward": 1.6640625,
      "reward_std": 0.2058405578136444,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.6640625,
      "step": 55
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1107.71875,
      "epoch": 0.7202572347266881,
      "grad_norm": 0.49433115124702454,
      "kl": 0.0011415481567382812,
      "learning_rate": 2.904938550483098e-07,
      "loss": 0.2208,
      "reward": 1.62890625,
      "reward_std": 0.18008438870310783,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.62890625,
      "step": 56
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1063.40625,
      "epoch": 0.7331189710610932,
      "grad_norm": 0.5648707151412964,
      "kl": 0.0015516281127929688,
      "learning_rate": 2.7402996237232757e-07,
      "loss": 0.1323,
      "reward": 1.58203125,
      "reward_std": 0.1835322491824627,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.58203125,
      "step": 57
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 842.5625,
      "epoch": 0.7459807073954984,
      "grad_norm": 0.6252850890159607,
      "kl": 0.0014705657958984375,
      "learning_rate": 2.5813805974175984e-07,
      "loss": 0.2248,
      "reward": 1.625,
      "reward_std": 0.18506385013461113,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.625,
      "step": 58
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 879.40625,
      "epoch": 0.7588424437299035,
      "grad_norm": 0.628139317035675,
      "kl": 0.0014867782592773438,
      "learning_rate": 2.4285108555160575e-07,
      "loss": 0.0506,
      "reward": 1.60546875,
      "reward_std": 0.18587078154087067,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.60546875,
      "step": 59
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1132.859375,
      "epoch": 0.7717041800643086,
      "grad_norm": 0.4942401647567749,
      "kl": 0.00098419189453125,
      "learning_rate": 2.2820072439029523e-07,
      "loss": 0.1978,
      "reward": 1.671875,
      "reward_std": 0.2286946102976799,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.671875,
      "step": 60
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1108.09375,
      "epoch": 0.7845659163987139,
      "grad_norm": 0.6833640336990356,
      "kl": 0.0009565353393554688,
      "learning_rate": 2.1421734136854153e-07,
      "loss": 0.1842,
      "reward": 1.625,
      "reward_std": 0.20673153176903725,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.625,
      "step": 61
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1164.8125,
      "epoch": 0.797427652733119,
      "grad_norm": 0.550803005695343,
      "kl": 0.0009765625,
      "learning_rate": 2.0092991918301106e-07,
      "loss": 0.0963,
      "reward": 1.578125,
      "reward_std": 0.18822026252746582,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.578125,
      "step": 62
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1222.84375,
      "epoch": 0.8102893890675241,
      "grad_norm": 0.47955459356307983,
      "kl": 0.0008296966552734375,
      "learning_rate": 1.883659980452598e-07,
      "loss": 0.092,
      "reward": 1.640625,
      "reward_std": 0.1938280612230301,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.640625,
      "step": 63
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1009.125,
      "epoch": 0.8231511254019293,
      "grad_norm": 0.6472100615501404,
      "kl": 0.0012683868408203125,
      "learning_rate": 1.765516186004387e-07,
      "loss": 0.1924,
      "reward": 1.671875,
      "reward_std": 0.20133696496486664,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.671875,
      "step": 64
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 997.796875,
      "epoch": 0.8360128617363344,
      "grad_norm": 0.6434372067451477,
      "kl": 0.00380706787109375,
      "learning_rate": 1.6551126795408015e-07,
      "loss": 0.2009,
      "reward": 1.66796875,
      "reward_std": 0.15900417417287827,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.66796875,
      "step": 65
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 942.125,
      "epoch": 0.8488745980707395,
      "grad_norm": 0.5874162912368774,
      "kl": 0.0014162063598632812,
      "learning_rate": 1.552678289188326e-07,
      "loss": 0.2217,
      "reward": 1.6484375,
      "reward_std": 0.18080222606658936,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.6484375,
      "step": 66
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 847.015625,
      "epoch": 0.8617363344051447,
      "grad_norm": 0.6690056920051575,
      "kl": 0.0014781951904296875,
      "learning_rate": 1.4584253258633681e-07,
      "loss": 0.1697,
      "reward": 1.6484375,
      "reward_std": 0.20908700302243233,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.6484375,
      "step": 67
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 967.515625,
      "epoch": 0.8745980707395499,
      "grad_norm": 0.6969915628433228,
      "kl": 0.0023241043090820312,
      "learning_rate": 1.3725491432254623e-07,
      "loss": 0.0535,
      "reward": 1.64453125,
      "reward_std": 0.12684179469943047,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.64453125,
      "step": 68
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1140.578125,
      "epoch": 0.887459807073955,
      "grad_norm": 0.5747631192207336,
      "kl": 0.0011663436889648438,
      "learning_rate": 1.2952277327769804e-07,
      "loss": 0.1446,
      "reward": 1.6171875,
      "reward_std": 0.1967878546565771,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.6171875,
      "step": 69
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1219.296875,
      "epoch": 0.9003215434083601,
      "grad_norm": 0.5590356588363647,
      "kl": 0.0020923614501953125,
      "learning_rate": 1.2266213549485637e-07,
      "loss": 0.1291,
      "reward": 1.58984375,
      "reward_std": 0.16909249685704708,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.58984375,
      "step": 70
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 871.53125,
      "epoch": 0.9131832797427653,
      "grad_norm": 0.6128804087638855,
      "kl": 0.002285003662109375,
      "learning_rate": 1.166872206934904e-07,
      "loss": 0.2013,
      "reward": 1.66015625,
      "reward_std": 0.18546020612120628,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.66015625,
      "step": 71
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 804.84375,
      "epoch": 0.9260450160771704,
      "grad_norm": 0.6195582151412964,
      "kl": 0.00205230712890625,
      "learning_rate": 1.1161041279693445e-07,
      "loss": 0.3434,
      "reward": 1.71875,
      "reward_std": 0.2264367826282978,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.71875,
      "step": 72
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1082.171875,
      "epoch": 0.9389067524115756,
      "grad_norm": 0.6192907691001892,
      "kl": 0.0029478073120117188,
      "learning_rate": 1.074422342648161e-07,
      "loss": 0.2337,
      "reward": 1.625,
      "reward_std": 0.2141621820628643,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.625,
      "step": 73
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 614.75,
      "epoch": 0.9517684887459807,
      "grad_norm": 0.75156170129776,
      "kl": 0.0022068023681640625,
      "learning_rate": 1.0419132428365116e-07,
      "loss": 0.1886,
      "reward": 1.796875,
      "reward_std": 0.18980605155229568,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.796875,
      "step": 74
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1100.421875,
      "epoch": 0.9646302250803859,
      "grad_norm": 0.4935654401779175,
      "kl": 0.0010290145874023438,
      "learning_rate": 1.0186442086081092e-07,
      "loss": 0.0693,
      "reward": 1.59375,
      "reward_std": 0.19873745739459991,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.59375,
      "step": 75
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1082.59375,
      "epoch": 0.977491961414791,
      "grad_norm": 0.4804929196834564,
      "kl": 0.0011501312255859375,
      "learning_rate": 1.004663468589726e-07,
      "loss": 0.1592,
      "reward": 1.64453125,
      "reward_std": 0.19965127855539322,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.64453125,
      "step": 76
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 972.4643249511719,
      "epoch": 0.9903536977491961,
      "grad_norm": 0.5502123236656189,
      "kl": 0.0015010833740234375,
      "learning_rate": 1e-07,
      "loss": 0.0737,
      "reward": 1.6640625,
      "reward_std": 0.212241068482399,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.6640625,
      "step": 77
    },
    {
      "epoch": 0.9903536977491961,
      "step": 77,
      "total_flos": 0.0,
      "train_loss": 0.13210956121238504,
      "train_runtime": 6757.7515,
      "train_samples_per_second": 0.046,
      "train_steps_per_second": 0.011
    }
  ],
  "logging_steps": 1,
  "max_steps": 77,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}